I saw the tear down on Jerry Rig Everything youtube channel and I was a bit underwhelmed by the cooling system, maybe that's why sustained speeds are not so great?
Six months ago I lost my job and after that I was fortunate enough to stumble upon a great website which literally saved me• I started working for them online and in a short time after I've started averaging 15k a month••• iｃash68.ｃｏM
1) Why iPhone 11 Pro Max's performance in 3DMark Physics is worse than iPhone XS's? 2) Do you agree that 3DMark is more RAM bound than GFXBench - and that's why iPhones perform worse than SD-smartphones with much RAM?
Because the Note9 is a big phone and the S9 is a smaller phone. The peak performance of those two units is identical, the sustained performance is determined by thermals and power.
Please, use updated results (PCMark work 2.0) of S10+(exynos) when you compare it to newer devices. I think your data is even android 9(not 10) for S10/S10+ devices. Here's what I get right now on my S10+ (exynos 9820, 12 GB version) updated to the latest official firmware, performance mode:
web browsing 2.0 my S10+/P: 9174 AT S10+(E9820): 7746 video editing my S10+/P: 6113 AT S10+(E9820): 5476 writing 2.0 my S10+/P: 9683 AT S10+(E9820): 9114 photo editing 2.0 my S10+/P: 20455 AT S10+(E9820): 11019 data manipulation my S10+/P: 6334 AT S10+(E9820): 5817 performance my S10+/P: 9321 AT S10+(E9820): 7531
See how much performance your result is missing? Photo editing 2.0 is almost 50% of its actual value... No point in comparing it with S20 ultra.
I have an iPhone 11 Pro Max and I can tell you without question that the phone has serious thermal throttling issues. As an example, I can play Dota Underlords at a fluid 60 fps until the phone warms up, and then things gradually drop to less than 15 fps. It is a very unpleasant experience.
Furthermore I am wondering if Apple is doing something to make the A13 appear better then it actually is. Don’t get me wrong, the iPhone 11 Pro Max and the A13 are both fantastic, but there are many odd quirks about both. In non gaming scenarios, the phone performs quite well, and I have no complaints thus far. However, in gaming scenarios the phone gets quite warm and throttles.
In essence you'd appreciate lower peak performance if it meant fewer throttling issues? I get the feeling lots of hardware is tends towards the running-too-hot spectrum, perhaps because it benchmarks better.
The thing is that for really intermittent stuff - UI animation, basically - it might be OK to have an absurdly high power ceiling since it's so short-term. Then again, you'd hope UI animation isn't a heavy load in the first place...
Yes, it seems like the ramp up and throttling are both too aggressive. If the phone were clocked a bit lower, I imagine there wouldn’t be throttling issues at all except in extreme environments. I wish Apple offered a bit more control over this.
I play a lot of heavy games on my 11 pro max, including fortnite and never experienced what you describe. From 60 to less than 15? :) You are playing only dota underlords? Because there are some games that misbehave like for example magic duels - it drains x3 fortnite high@60fps and heats the phone like no tomorrow. This is literally the only game I ever had issue with and I got around atleast 30 games and 15-20 from them are one of the most demanding on the app store.
3DMark Physics is essentially a test of (number of cores)*(frequency). It's not really a test of the GPU, and the way it tests a CPU is mostly not representative of most code.
Apple has looked worse on it than any other benchmark since smartphones started for precisely this reason -- Apple ships with fewer cores and frequently clocked lower than Android.
All benchmarks are ideal scenarios. Real life code never matched the theoretical advantages of DX12 vs DX11 for example. Apple looks worse in this test because it's also about thermals. Apple's high performance SoC's Achilles' heel is throttling under sustained workloads.
User eek2121 describes above how the mighty pro max drops to unplayable 15fps becacuse it runs too hot.
orrr he have a defective device that is misbehaving, or he have a totally f up software that cries for clear reinstall without any backup restores. One user is one user + you can't even prove what he says is true. You can read Andrei's review and see how the pro max behaves when stressed.
If you're not even interested in the difference between 3DMark PHYSICS and 3DMark then you don't have anything technical to contribute to this discussion. The PHYSICS benchmark has very different characteristics from the rest. If you want to use it as a tool to understand SoCs, you ought to start by appreciating that point.
It's not a cheering match! I'm trying to explain the characteristic of the benchmark. That's a technical statement, it has nothing to do with who prefers what company. It's no different from pointing out that a benchmark like Lua is going to stress indirect branches, or that mcf depends strongly on memory latency.
Look at the historical values of 3DMark Physics across a wide variety of devices (Apple or otherwise). Obsessing/cheering over whether the iPhone Pro Max is throttling or not blinds you to the larger issues of how the benchmark behaves generically.
I was not talking about physics only. The GPU version of 3dmark is the same. Check out the sustained numbers and note how the delta is so much higher on the iphone.
Spec on this site is ran on an active cooled setup. Otherwise no phone, android or ios, can run it without throttling massively or crashing. In this aspect the 3dmark test is a more realistic, phone in hand scenario.
Or do you see that? I see in: physics: s9 (2591/1513) Note 9 (2659/2227) Graphics: s9 (3724/2179) Note 9 (3832/3639) The performances are very close, for the second values it is normal it is about sustained performance which therefore depends by thermal. same for the iphone. Physics: 11 pro max (3527/2642) XS (3087/2910) Graphics: 11 pro max (8677/5944) XS (4439/3929). Scores are consistent
Any preliminary thoughts on Exynos 990 vs. SD865? Being in Europe it's always one of the main considerations with these devices. I'm not hearing good things about the battery on the Exynos 990 ones at all.
Can't wait for the Exynos comparison. Any word on the next Exynos? Surely it would be Cortex-A78 now that SARC team is gone, or would it be M5 again? The late 2021 model should be using RDNA+ 2.5 years after the announcement. So the next Exynos could be the only Exynos flagship using ARM CPU+GPU IP in a long while.
I have it hard to believe that the QRD865 performs barely as well as an iPhone 7, and half of the iPhone 11. Considering that these are "WebView" benchmarks, the more logical explanation is that the iPhone browser performs better than the Android one in these tests. As a simplistic example, imagine that the JavaScript engine runs at 30 fps on iPhone and 60fps on Android, that alone would make a huge difference.
Is there a pure CPU benchmark where the same instructions are run on these CPUs so that we can have a comparison with fewer variables?
If I understand the SPEC benchmark correctly, the QRD865 consumes about half of the power of the iPhone 11 to produce about half of its performance. Correct?
That would mean that the iPhone's CPU is designed to draw a lot more power... I wonder why Qualcomm doesn't have a CPU with similar design characteristics. Maybe the monetary benefits don't justify it?
...though would you want that? I can't imagine intentionally damaging battery life expectancy by so much for so little perf gain. But given the otherwise excellent A13, imagine what they could do if they aimed a little more at sustained perf and battery life, and less at instant space-heater... I'm curious how the existing Low Power Mode compares, although that disables various other bits of functionality you might not want to lose.
Yes. I would love to see the Low Power Mode in the SPEC sheets as well in order to compare it to the "middle cores" in the 865/855. IIRC, Andrei posted a voltage curve in its review, and it seems like the upper 200Mhz consumes a lot.
My understanding is that the little cores in the A13 are significantly better than the little cores in the SD865. This allows Apple to keep more tasks on those very efficient cores, which is good for battery life. This gives them the headroom to go a bit nuts on the big cores.
Though, I would also be interested to see an A13 limited to 70% on the big cores. I imagine it would still have plenty of power and would be much better for battery life. Of course, given the data Apple has about how people use their phones, it's probably already optimized for the best overall balance.
Actually the power is 2.5x lower for a performance about 1.5x lower. For power measurement, look at the Watt figure on the left chart (first one before the comma).
A bigger die like Apple's would be more expensive to produce and hence would have few takers in the android side. The discrete modem this year alone led to such a big rise in flagship prices. Imagine what a large CPU + discrete modem would have done.
Qualcomm gets their CPU designs from ARM as well. So if ARM doesn't make a big CPU, Qualcomm just has to take what ARM has or develop their own (which is very hard).
Essentially all you have to know is the X performs better than Y in certain benchmarks. Now how accurate those benchmarks are is another discussion. Everybody can see than the iphone 7 doesn't feel faster in day to day usage than a lot of mid-range android phones.
Edit: I wish you would be more specific than just using adjectives like "superior". I look up to this website for educational content. It would also be great to have links after each benchmark to a FAQ that explains how we should understand it (so, for the WebView one, the influence of the JavaScript engine is a big disclaimer to have).
That performance section covers "system performance" as a whole of hardware and software combined, while yes JS engine perf matters, it shouldn't matter in this context of overall device perf.
As for Apple, you can read more details in the dedicated reviews of those phones, I can't really go into more specifics in the comment section other than summing up that it's "superior":
There is also the issue of testing mobile phone platforms in general, and I try to make sure everyone is aware of this: Apple phones use different languages and frameworks. They also run a different operating system. This may cause some of the performance differences we see.
Note that I’m not sandbagging on anyone, I am just encouraging an objective view of things.
"Apple's CPUs are just far superior, see the SPEC scores for an even comparison of workload."
Yes, but not to this degree, right? Speaking specifically about this browsing benchmark that shows iPhone 7 being superior to Galaxy S20, it absolutely does NOT translate into real-life experience. Galaxy S20 offers a vastly superior browsing experience to iPhone 7.
My point is that these benchmarks would show the SD 710 and much much weaker than an iphone 7 but in real world phone usage it isn't. And SD 710 was from the beginning a mid-range chip and it's found even now in phones that are much cheaper then an iphone 7.
It'll be an enormous rectangle with almost all screen at the front and delicate rounded edges that prevent you from using a screen protector. Everything is just copy and paste these days, with very anti-consumer designs
Places like the the Verge cover more the look of the phone. This website covers more the performance characteristics of the phone. Why is this bad? Also, the title said this: "Quick Performance Preview". It did not say general review.
The more I think about it, the more I feel the numbers don't add up. ARM claims no efficiency gain at the same frequency, and TSMC only promised 10% power reduction, yet perf/W increased by 40%, perf/J a whopping 66%, fp even more.
In this generation they have double L3, lower memory latency and higher memory bandwidth. So it is not a apples-to-apples comparison :) S865 will obviously be better
That doesn't explain a whopping 30% increase in efficiency. Doubling L3 has diminishing benefit, I doubt it can improve performance by 1% considering 2MB is already very big for single core. And single core is obviously not bandwidth bound. You can put 16MB L3 and LPDDR5-3200 on it, the single core performance won't increase even 5%.
That's impossible. No amount of cache and memory bandwidth can improve A77's efficiency by 30 to 80 percent given the baseline is 2MB L3$ + 3MB Sys$ for a single 2-watt CPU core.
Otherwise Ryzen 2600X would have the same advantage over 2500X, yet it definitely doesn't.
Increasing caches has a huge impact on efficiency - on memory intensive benchmarks you will now be able to run everything from L3 rather than waiting forever on DRAM. A quadrupling of caches would significantly reduce the gap with Apple, gaining 20-30% is easy *if* you are willing to spend the area.
Wrong - a larger L3 makes a huge difference, 32% when using a 32MB L3. From the Graviton2 article:
"Compared to a mobile Cortex-A76 such as in the Kirin 990 (which is the best A76 implementation out there), the resulting IPC is 32% better for the Graviton2 in SPECint2006, and 10% better for SPECfp2006. This goes to show what kind of a massive difference the memory subsystem can have on a system that is otherwise similar in terms of the CPU microarchitecture."
Generally, doubling cache size does not help performance much unless the cache is far too small already. The memory subsystem is about far more than cache. Memory controller design, clock syncing vs async, trace distances, set associativity, and many other factors come into play. Look at any example in history of cache doubling on a given architecture. You will not see a 30% gain, certainly not an 80% gain.
You're muddling ipc uplift vs power efficiency gain. Yes, on a spec, sub system won't buy you much save in mcf type workloads on raw perf. But you're overlooking the dram effect. On die cache reduces traffic to dram, which for a mobile soc, is exponentially more expensive to access both thermally and energy wise than any horseplay on the soc. The result is real. And even Apple knows this, their massive caches and resultant perf/W are not a mistake. Dram is your mobile enemy for battery life.
In SpecInt2k6 there are tests that are very sensitive to memory latency and others which are very sensitive to memory bandwidth. So you can increase perf a lot. Doubling the L3 cache gives you a lot more perf (Spec data set size is way bigger than 2MB) and also saves on power because you go less often to DDR memory which is crazy high power.
865 processor sucks versus Apple. The iPhone 11 Pro is whopping 100% faster than S20 on Speedometer2 and JetSctream2 yet Samsung insists in equal prices... Hahahaha
The S20 is a true end to end flagship with a top of the line OLED screen and camera system. It's comparable to the iphone 11 Pro not the plain 11. iPhone 11 basically has the same mediocre screen found in the XR, exactly the same screen, apple mostly only changed the SOC and camera system.
Completely stupid reasoning. 1) His tests are developed by webkit (so we imagine safari advantage) 2) The web tests cannot be compared between different platforms and are used to demonstrate the performance of a browser within the same bone / hardware (They are mainly used to demonstrate the performance of browsers not those of the device and generally developers benefit their browsers For example, in jetstream (develop by webkit) safari is 2x better than firefox, conversely kraken (develop by firefox) firefox is 2x better. 3) To compare, you should use pcmark for example which tests the browsing performance and not the browser
It doesn't load faster. Also in general flagship Android phones have better LTE and Wifi performance, that matters more when it comes to webpage loading.
Fact is you can't produce any evidence showing your iPhone loads webpages faster. All I see are benchmarks that don't translate into real life experience.
These synthetic tests are boring. I'm not going to buy a device to play this. Could you test it on some real-use scene ? Test on heavy games and emulators like Play/Dolphin/DamonPS2.
I generally don't like apple products, but I think this time Apple got it right, a 5.8-inch phone(Iphone 11pro) fits perfectly in the hand. These things of almost 7inch are very uncomfortable to use with just one hand. e.e
The casually included MT6885Z benchmark results are really sticking out to me. Are you going to review the Redmi K30 or a similar device? If so, I'm really looking forward to it.
Following the shambles of S.LSI's previous "adventures" I'm kind of curious as to how SARC's M5 cores end up performing compared to QC's products, but considering the fact that they've scrapped it entirely along with the fact that they're not shipping the Exynos version in their own market (South Korea) along with the struggles of Samsung's Foundry business, I highly doubt that Samsung will be able to be competitive for a very long time...
Anyway: I appreciate your work and hope that you'll continue to deliver excellent content like this in the future.
I also came here to comment on the inclusion of the Dimensity 1000L in the SPEC numbers. Are we getting a full review and/or analysis of a device with this SoC?
Also interested, it appears that the supposed 2.6Ghz A77 is performing worse than the 2.4Ghz A77 in SD865, the 2.86Ghz A76 in Kirin 990, and then 2.84Ghz A76 in SD855... which would point to some deficiencies in how Mediatek implemented their memory hierarchy beyond the L1 in their SoC?
Keep in mind that the A77 cores in the Dimensity 1000L are only running with up to 2.2 GHz (it's a slightly downrated version of the Dimensity 1000). If you adjust the SPECSpeed for clocks, the A77 cores should fall in line with the other A77-based SoCs, assuming that there are no bottlenecks that would affect the performance scaling negatively.
IMO I believe for my $ anything considered "flagship" class, should dang well include 3.5mm jack, BT / wireless is "cool" and all that, latency wise however (not to mention connectivity options) 3.5mm jack still has so much more going for it.
BT/Wireless has a place, no doubt, but gosh dang, over $1100 phone (above and beyond most full out laptops, which tend to carry audio jacks LOL) then you have to go and buy "respectable" quality earbuds on top of this usually in the $50+ as well.
the world is made of money and we like trashing things every year or 2 "cause that is how we roll"
Oppio (or whatever they called, I typo I know) and some others have very near same spec (not Exynos chip of course) less pricey and many still carry the venerable audio/aux jack not make stupid excuses about not having it either
TL;DR: in normal mode which will be used by 99% of customers, without the battery-killing "Performance" option enabled, S20 Ultra is about the same as S10+, and even with the option still trails last year iPhones by a whole lot.
Yes, I just thought the same. I just focused the SD865/60Hz scores, and ignored the 120Hz and P mode.
The reason being even if I or any sane person considering buying S20 series device, its either going to be S20 with its 4000mah battery or the S20+ with 4500mah one.
So when I think of how fast is it from S10+ with the same battery size as S20, the numbers are not interesting at all.
You also need to keep in mind that S20 Ultra is a Huge phone, with lots of body space to cool it down, so the performance numbers might be different from the smaller models, and whatever battery/power the new architecture of SD865 is going to save is more probably going to be wasted by the external modem.
Yeah I'm most interested in the S20 normal results, not the + or Ultra. It is already an extra $300 over the iPhone 11, I'm not really interested in the phone that costs double. I'm not interested in the 11 Pro Max etc. either.
Reviewers love to focus on superficial stuff like specs and benchmarks.
Ever notice how flagship phones lag, even on release? The software is EXTREMELY badly written. A lot of these apps should be using maximum 1 megabyte of RAM.. Instead they're loading gigabytes of assets and libraries, makes no sense at all, but makes it much easier on the developers. I wish Terry A. Davis had written a modern smartphone OS. Multi-tasking support should be better, and app-switching should be instantaneous. Even Microsoft Windows can do that.
Oh, and also, No front speakers = no buy. No headphone jack = no buy. And unless it performs miracles (it doesn't), $999 = no buy.
That's why I picked up my first iPhone since the iPhone 4 this week. Cheap, fast, works (yet sadly filled with stupid Apple decisions, but ones I can live with). I want iMessage for the browser, and T9 dialing, basic stuff that Apple refuses to do out of pigheadedness. But it works, battery life (stand by time in particular) is amazing. My less than one year old flagship Android phone, it's Android auto stopped working (the voice assistant), and Google has known about the bugs for 6 months now and hasn't fixed them. Bought the iPhone for Carplay, which is great. Only problem is I use all google services, so I have to append "on Youtube music" to every request, which is annoying.
Android and Apple both need to add a default music player setting option, for god's sake.
I'm a Brit and have had the Exynos 990 S20 Ultra since Monday - really impressed - previous phone being a Pixel 2 XL. Completely agree with the review that the 120Hz mode is a significant step forward - I don't even mind it being a lower resolution. I guess I'm coming forward 3 CPU generations (S835) even though it's less than 2.5yrs since I got the P2XL. Completely expecting the S865 to be superior to the E990 but it's still overall a very good phone. Ironically not had much chance to use the camera yet, but speakers are infinitely better than the P2.
You forget one important test 4K HDR YouTube video with full resolution of the S20 ultra cause S10 Exynos and S10 snapdragon 855 lags if you watch these kind of extreme resolution videos and throttling so fast snapdragon throttle even worst than Exyons.
Finally some generational improvements like 820 and 835 (efficiency), I think this is a good platform to buy and get it going for a long time since the battery consumption of these planned obsolescence devices will be at-least good enough to keep them going for 3+ years instead of just 2.
Coming to SPEC, yeah as usual. Apple's iPhone 7 is defeating this, but one simple question, will the consumer market, Including the supply and demand with respect to the ROI and performance (meaning people's usage patterns and their satisfaction rate) so does these benchmarks translate to the pricing and the user experience ? to put it simply, an iPhone 8 or X as shown in SPEC, will they be able to beat an S10 or S20 experience ? I don't think so at all..
Next up, Samsung S20 series phones are ugly and looks like trash, their camera module is so big and thick and it's an eye sore comparing it to the Note 8,9, S10 this is a huge downgrade, when DJ Koh left Samsung, or since the rumors were coming, I speculated changes but not for worse.
- Lack of SpO2 sensor in Samsung health app now, it's just a pedometer now. - Removal of 3.5mm jack, what the fuck is this ? This phone is right up at over $1000 without tax and it lacks basic stuff like that for no reason except for greed, their stupid Galaxy Buds do not even have AptX or AptXLL and they use cheap shit Balanced Armatures and demanding over $100 for that ANC, absolute rip off when you see their guts sporting coin cells which could be slot loaded or such but we have another planned obsolescence gadget. AKG is trash, AKG N5005 is shit, they have huge regression in the whole QC, Sound, K3003i was great but this is nothing. Shows in their Galaxy Buds and inbox set. - Insane price for the base version as well. - Seeing the camera, they removed that Dual Pixel PDAF and got focusing issues for some nonabinning bullshit. - Still having a hole despite showing off their Under Screen Camera in 2018, in China. - Their Knox chip, I wonder Exynos will have bootloader unlock or not this time and how hard is it. - No stereo speakers which are FF, I know their sound is great on Note10 but it should have been done by now. - 120Hz for bragging rights, downgraded the display resolution to FHD. It's a shame when their OP7P panel does it at 1440P, they couldn't give the base version at 5K mah ? Zenfone 6 did it last year, LG V60 is doing it this year..
Mr. Frumusanu, does Manhattan support Vulkan on Android? They don't advertise Vulkan support on Android but they do support Metal on ios. Roughly the same question for T-Rex.
What are your Financial needs? We give out loans from a minimum of $2,000.00 to a maximum of $500,000,000.00 with comfortable duration that ranges from 1 to 30 years at a very reduced interest rate of 3%. Do you need a business loan? Do you need a personal loan? Do you want to buy a car? Do you want to refinance? Do you need a mortgage loan? Do you need a huge capital to start off your business proposal or expansion? Have you lost hope and you think there is no way out, and your financial burdens still persists? Please do not hesitate to contact us for possible business co-operation.
Our services include the following: *Refinancing Loans * Car Loan *Truck Loans * Home Loan * Mortgage Loan * Debt Consolidation Loan * Business Loan [secure and unsecured] * Personal Loan [secure and unsecured] * Students Loan and so many others.
Sincerely, Dr. Washington Ray. (Credit Officer) Text/Call +1(917)983-0070 ([email protected])
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
117 Comments
Back to Article
Tigran - Friday, March 13, 2020 - link
As far as I remember Samsung's top Snapdragon smartphones were always worst in GFXBench.Tigran - Friday, March 13, 2020 - link
At least when it's about sustained performance.NICOXIS - Friday, March 13, 2020 - link
I saw the tear down on Jerry Rig Everything youtube channel and I was a bit underwhelmed by the cooling system, maybe that's why sustained speeds are not so great?dianajmclean6 - Monday, March 23, 2020 - link
Six months ago I lost my job and after that I was fortunate enough to stumble upon a great website which literally saved me• I started working for them online and in a short time after I've started averaging 15k a month••• iｃash68.ｃｏMLuckky26 - Tuesday, April 7, 2020 - link
is that why now they've started launching phones with Mediatek chipsets, lolTigran - Friday, March 13, 2020 - link
1) Why iPhone 11 Pro Max's performance in 3DMark Physics is worse than iPhone XS's?2) Do you agree that 3DMark is more RAM bound than GFXBench - and that's why iPhones perform worse than SD-smartphones with much RAM?
Andrei Frumusanu - Friday, March 13, 2020 - link
1) A13 runs hotter, hence more throttling2) No. It's more compute bound, probably some shader combination on which Adreno does better.
Tigran - Friday, March 13, 2020 - link
2) Then why Galaxy Note 9 (6 GB) outperforms Galaxy S9 (4 GB), both E9810?Physics: 2227 vs 1413
Graphics: 3639 vs 2179
https://www.anandtech.com/show/14072/the-samsung-g...
Andrei Frumusanu - Friday, March 13, 2020 - link
Because the Note9 is a big phone and the S9 is a smaller phone. The peak performance of those two units is identical, the sustained performance is determined by thermals and power.Tigran - Friday, March 13, 2020 - link
Agree. Thanks! Just wonder were I've read about 3DMark being RAM bound...dad_at - Sunday, March 15, 2020 - link
Please, use updated results (PCMark work 2.0) of S10+(exynos) when you compare it to newer devices. I think your data is even android 9(not 10) for S10/S10+ devices. Here's what I get right now on my S10+ (exynos 9820, 12 GB version) updated to the latest official firmware, performance mode:web browsing 2.0 my S10+/P: 9174 AT S10+(E9820): 7746
video editing my S10+/P: 6113 AT S10+(E9820): 5476
writing 2.0 my S10+/P: 9683 AT S10+(E9820): 9114
photo editing 2.0 my S10+/P: 20455 AT S10+(E9820): 11019
data manipulation my S10+/P: 6334 AT S10+(E9820): 5817
performance my S10+/P: 9321 AT S10+(E9820): 7531
See how much performance your result is missing? Photo editing 2.0 is almost 50% of its actual value... No point in comparing it with S20 ultra.
eek2121 - Friday, March 13, 2020 - link
I have an iPhone 11 Pro Max and I can tell you without question that the phone has serious thermal throttling issues. As an example, I can play Dota Underlords at a fluid 60 fps until the phone warms up, and then things gradually drop to less than 15 fps. It is a very unpleasant experience.Furthermore I am wondering if Apple is doing something to make the A13 appear better then it actually is. Don’t get me wrong, the iPhone 11 Pro Max and the A13 are both fantastic, but there are many odd quirks about both. In non gaming scenarios, the phone performs quite well, and I have no complaints thus far. However, in gaming scenarios the phone gets quite warm and throttles.
emn13 - Friday, March 13, 2020 - link
In essence you'd appreciate lower peak performance if it meant fewer throttling issues? I get the feeling lots of hardware is tends towards the running-too-hot spectrum, perhaps because it benchmarks better.The thing is that for really intermittent stuff - UI animation, basically - it might be OK to have an absurdly high power ceiling since it's so short-term. Then again, you'd hope UI animation isn't a heavy load in the first place...
eek2121 - Friday, March 13, 2020 - link
Yes, it seems like the ramp up and throttling are both too aggressive. If the phone were clocked a bit lower, I imagine there wouldn’t be throttling issues at all except in extreme environments. I wish Apple offered a bit more control over this.flyingpants265 - Friday, March 13, 2020 - link
We should worry about nonsense like "UI animations" AFTER the devices are already fast, which they're not.cha0z_ - Saturday, March 14, 2020 - link
I play a lot of heavy games on my 11 pro max, including fortnite and never experienced what you describe. From 60 to less than 15? :)You are playing only dota underlords? Because there are some games that misbehave like for example magic duels - it drains x3 fortnite high@60fps and heats the phone like no tomorrow. This is literally the only game I ever had issue with and I got around atleast 30 games and 15-20 from them are one of the most demanding on the app store.
GeorgiaPortus - Saturday, March 21, 2020 - link
Pass your next certififcation exam fast through <a href="https://www.certsofficial.com/">Certs Official</a> to learn more and get 100% result.name99 - Friday, March 13, 2020 - link
3DMark Physics is essentially a test of (number of cores)*(frequency). It's not really a test of the GPU, and the way it tests a CPU is mostly not representative of most code.Apple has looked worse on it than any other benchmark since smartphones started for precisely this reason -- Apple ships with fewer cores and frequently clocked lower than Android.
https://www.anandtech.com/show/7335/the-iphone-5s-...
id4andrei - Saturday, March 14, 2020 - link
All benchmarks are ideal scenarios. Real life code never matched the theoretical advantages of DX12 vs DX11 for example. Apple looks worse in this test because it's also about thermals. Apple's high performance SoC's Achilles' heel is throttling under sustained workloads.User eek2121 describes above how the mighty pro max drops to unplayable 15fps becacuse it runs too hot.
cha0z_ - Saturday, March 14, 2020 - link
orrr he have a defective device that is misbehaving, or he have a totally f up software that cries for clear reinstall without any backup restores. One user is one user + you can't even prove what he says is true. You can read Andrei's review and see how the pro max behaves when stressed.id4andrei - Sunday, March 15, 2020 - link
I've read it. 3dmark is all about thermals and Snapdragon is thermally more sustainable.iphonebestgamephone - Monday, March 16, 2020 - link
Then why does the kirin 990 score higher than sd855+? Better heatpipes on the mate 30 pro?name99 - Monday, March 16, 2020 - link
If you're not even interested in the difference between 3DMark PHYSICS and 3DMark then you don't have anything technical to contribute to this discussion.The PHYSICS benchmark has very different characteristics from the rest. If you want to use it as a tool to understand SoCs, you ought to start by appreciating that point.
name99 - Monday, March 16, 2020 - link
It's not a cheering match!I'm trying to explain the characteristic of the benchmark. That's a technical statement, it has nothing to do with who prefers what company. It's no different from pointing out that a benchmark like Lua is going to stress indirect branches, or that mcf depends strongly on memory latency.
Look at the historical values of 3DMark Physics across a wide variety of devices (Apple or otherwise). Obsessing/cheering over whether the iPhone Pro Max is throttling or not blinds you to the larger issues of how the benchmark behaves generically.
id4andrei - Monday, March 16, 2020 - link
I was not talking about physics only. The GPU version of 3dmark is the same. Check out the sustained numbers and note how the delta is so much higher on the iphone.Spec on this site is ran on an active cooled setup. Otherwise no phone, android or ios, can run it without throttling massively or crashing. In this aspect the 3dmark test is a more realistic, phone in hand scenario.
iphonebestgamephone - Tuesday, March 17, 2020 - link
You were replying to a comment which specifically mentioned the phyiscs test. Why is sd worse than the kirin?doungmli - Saturday, March 14, 2020 - link
Or do you see that? I see in:physics:
s9 (2591/1513)
Note 9 (2659/2227)
Graphics:
s9 (3724/2179)
Note 9 (3832/3639)
The performances are very close, for the second values it is normal it is about sustained performance which therefore depends by thermal.
same for the iphone.
Physics:
11 pro max (3527/2642)
XS (3087/2910)
Graphics:
11 pro max (8677/5944)
XS (4439/3929).
Scores are consistent
jaju123 - Friday, March 13, 2020 - link
Any preliminary thoughts on Exynos 990 vs. SD865? Being in Europe it's always one of the main considerations with these devices. I'm not hearing good things about the battery on the Exynos 990 ones at all.Andrei Frumusanu - Friday, March 13, 2020 - link
I'm not getting the 990 till tomorrow or Monday, not much to say at this point.Luckky26 - Tuesday, April 7, 2020 - link
Even the snapdragon 865 is expected to show some issues due to a non-integrated 5G modem which apparently takes up more spacedotjaz - Friday, March 13, 2020 - link
Can't wait for the Exynos comparison.Any word on the next Exynos? Surely it would be Cortex-A78 now that SARC team is gone, or would it be M5 again? The late 2021 model should be using RDNA+ 2.5 years after the announcement. So the next Exynos could be the only Exynos flagship using ARM CPU+GPU IP in a long while.
Andrei Frumusanu - Friday, March 13, 2020 - link
Next one is A78, yes.eek2121 - Friday, March 13, 2020 - link
I am still quite baffled why Samsung didn’t go all in instead of folding. IMO they should have brought the Exynos to all markets.dotjaz - Friday, March 13, 2020 - link
They are not folding Exynos, just Exynos M cores.BedfordTim - Friday, March 13, 2020 - link
Any ideas why data manipulation, and other benchmarks are dependent on display frame rate?Andrei Frumusanu - Friday, March 13, 2020 - link
They're hitting V-sync, higher V-sync = higher scores.emn13 - Friday, March 13, 2020 - link
Which also means the value of the test is dubious to start with.eek2121 - Friday, March 13, 2020 - link
I wasn’t going to go down that road, but I agree. Without being able to disable v-sync, the value of the tests are limited.SafraneIt - Friday, March 13, 2020 - link
I have it hard to believe that the QRD865 performs barely as well as an iPhone 7, and half of the iPhone 11.Considering that these are "WebView" benchmarks, the more logical explanation is that the iPhone browser performs better than the Android one in these tests.
As a simplistic example, imagine that the JavaScript engine runs at 30 fps on iPhone and 60fps on Android, that alone would make a huge difference.
Is there a pure CPU benchmark where the same instructions are run on these CPUs so that we can have a comparison with fewer variables?
Andrei Frumusanu - Friday, March 13, 2020 - link
Apple's CPUs are just far superior, see the SPEC scores for an even comparison of workload.SafraneIt - Friday, March 13, 2020 - link
If I understand the SPEC benchmark correctly, the QRD865 consumes about half of the power of the iPhone 11 to produce about half of its performance. Correct?That would mean that the iPhone's CPU is designed to draw a lot more power...
I wonder why Qualcomm doesn't have a CPU with similar design characteristics. Maybe the monetary benefits don't justify it?
Andrei Frumusanu - Friday, March 13, 2020 - link
2.5x/2.13x the power for 1.57x/1.37x the performance. Arm will only have a similar "high power" CPU in 2021+.SafraneIt - Friday, March 13, 2020 - link
Thank you for the details.emn13 - Friday, March 13, 2020 - link
...though would you want that? I can't imagine intentionally damaging battery life expectancy by so much for so little perf gain. But given the otherwise excellent A13, imagine what they could do if they aimed a little more at sustained perf and battery life, and less at instant space-heater... I'm curious how the existing Low Power Mode compares, although that disables various other bits of functionality you might not want to lose.dudedud - Friday, March 13, 2020 - link
Yes. I would love to see the Low Power Mode in the SPEC sheets as well in order to compare it to the "middle cores" in the 865/855. IIRC, Andrei posted a voltage curve in its review, and it seems like the upper 200Mhz consumes a lot.cfenton - Saturday, March 14, 2020 - link
My understanding is that the little cores in the A13 are significantly better than the little cores in the SD865. This allows Apple to keep more tasks on those very efficient cores, which is good for battery life. This gives them the headroom to go a bit nuts on the big cores.Though, I would also be interested to see an A13 limited to 70% on the big cores. I imagine it would still have plenty of power and would be much better for battery life. Of course, given the data Apple has about how people use their phones, it's probably already optimized for the best overall balance.
arayoflight - Friday, March 13, 2020 - link
Actually the power is 2.5x lower for a performance about 1.5x lower. For power measurement, look at the Watt figure on the left chart (first one before the comma).A bigger die like Apple's would be more expensive to produce and hence would have few takers in the android side. The discrete modem this year alone led to such a big rise in flagship prices. Imagine what a large CPU + discrete modem would have done.
Qualcomm gets their CPU designs from ARM as well. So if ARM doesn't make a big CPU, Qualcomm just has to take what ARM has or develop their own (which is very hard).
Nicon0s - Monday, March 23, 2020 - link
Essentially all you have to know is the X performs better than Y in certain benchmarks.Now how accurate those benchmarks are is another discussion.
Everybody can see than the iphone 7 doesn't feel faster in day to day usage than a lot of mid-range android phones.
SafraneIt - Friday, March 13, 2020 - link
Edit: I wish you would be more specific than just using adjectives like "superior". I look up to this website for educational content.It would also be great to have links after each benchmark to a FAQ that explains how we should understand it (so, for the WebView one, the influence of the JavaScript engine is a big disclaimer to have).
Andrei Frumusanu - Friday, March 13, 2020 - link
That performance section covers "system performance" as a whole of hardware and software combined, while yes JS engine perf matters, it shouldn't matter in this context of overall device perf.As for Apple, you can read more details in the dedicated reviews of those phones, I can't really go into more specifics in the comment section other than summing up that it's "superior":
https://www.anandtech.com/show/13392/the-iphone-xs...
https://www.anandtech.com/show/14892/the-apple-iph...
eek2121 - Friday, March 13, 2020 - link
There is also the issue of testing mobile phone platforms in general, and I try to make sure everyone is aware of this: Apple phones use different languages and frameworks. They also run a different operating system. This may cause some of the performance differences we see.Note that I’m not sandbagging on anyone, I am just encouraging an objective view of things.
arsjum - Saturday, March 14, 2020 - link
"Apple's CPUs are just far superior, see the SPEC scores for an even comparison of workload."Yes, but not to this degree, right? Speaking specifically about this browsing benchmark that shows iPhone 7 being superior to Galaxy S20, it absolutely does NOT translate into real-life experience. Galaxy S20 offers a vastly superior browsing experience to iPhone 7.
Nicon0s - Monday, March 16, 2020 - link
Yeah far superior in SPEC2006. Big deal.In real life iPhones 7 doesn't feel any faster than a mid-range Android phone with something like an SD 710.
Wolf92 - Tuesday, March 17, 2020 - link
You're comparing a phone released in september of 2016 with a SoC released in may of 2018. What exactly is your point?Nicon0s - Wednesday, March 18, 2020 - link
My point is that these benchmarks would show the SD 710 and much much weaker than an iphone 7 but in real world phone usage it isn't.And SD 710 was from the beginning a mid-range chip and it's found even now in phones that are much cheaper then an iphone 7.
sonny73n - Friday, March 13, 2020 - link
Is the phone so plain that Anandtech only shows 1 picture of it?yetanotherhuman - Friday, March 13, 2020 - link
It'll be an enormous rectangle with almost all screen at the front and delicate rounded edges that prevent you from using a screen protector. Everything is just copy and paste these days, with very anti-consumer designsnicmonson - Friday, March 13, 2020 - link
Places like the the Verge cover more the look of the phone. This website covers more the performance characteristics of the phone. Why is this bad? Also, the title said this: "Quick Performance Preview". It did not say general review.BushLin - Friday, March 13, 2020 - link
It's ok, you will see a heavy focus on the size of the bezels when arstechnica review the handset.SafraneIt - Friday, March 13, 2020 - link
OT: Will you review the Surface Pro X? Or at least benchmark its "Microsoft SQ1" CPU?It is probably the highest power CPU available from Snapdragon.
masimilianzo - Friday, March 13, 2020 - link
Hi Andrei,Just a small request..going forward could you include a disclaimer about your methodology for power/energy estimation?
Andrei Frumusanu - Friday, March 13, 2020 - link
I'm outright measuring the power via different methods, it's not an estimation.masimilianzo - Friday, March 13, 2020 - link
Good, see your point.Still could you share the different methods you are using?
dotjaz - Friday, March 13, 2020 - link
The more I think about it, the more I feel the numbers don't add up. ARM claims no efficiency gain at the same frequency, and TSMC only promised 10% power reduction, yet perf/W increased by 40%, perf/J a whopping 66%, fp even more.masimilianzo - Friday, March 13, 2020 - link
In this generation they have double L3, lower memory latency and higher memory bandwidth.So it is not a apples-to-apples comparison :)
S865 will obviously be better
dotjaz - Friday, March 13, 2020 - link
That doesn't explain a whopping 30% increase in efficiency. Doubling L3 has diminishing benefit, I doubt it can improve performance by 1% considering 2MB is already very big for single core.And single core is obviously not bandwidth bound.
You can put 16MB L3 and LPDDR5-3200 on it, the single core performance won't increase even 5%.
Andrei Frumusanu - Friday, March 13, 2020 - link
You're just wrong there, both those aspects will have huge impacts on ST perf and efficiency.dotjaz - Friday, March 13, 2020 - link
That's impossible. No amount of cache and memory bandwidth can improve A77's efficiency by 30 to 80 percent given the baseline is 2MB L3$ + 3MB Sys$ for a single 2-watt CPU core.Otherwise Ryzen 2600X would have the same advantage over 2500X, yet it definitely doesn't.
Wilco1 - Friday, March 13, 2020 - link
Increasing caches has a huge impact on efficiency - on memory intensive benchmarks you will now be able to run everything from L3 rather than waiting forever on DRAM. A quadrupling of caches would significantly reduce the gap with Apple, gaining 20-30% is easy *if* you are willing to spend the area.dotjaz - Friday, March 13, 2020 - link
So then, you can explain why Thunder didn't have that with doubled L2$. Not even close to 66% per J improvement.dotjaz - Friday, March 13, 2020 - link
A12 didn't have similar improvement over A11, that's with N7 vs N10. Thunder didn't have the same effect with doubled L2 either.Don't try to convince anyone that mere LPDDR5 vs 4x can cause such a big difference for single thread, we both know that's not even remotely possible.
Wilco1 - Friday, March 13, 2020 - link
Wrong - a larger L3 makes a huge difference, 32% when using a 32MB L3. From the Graviton2 article:"Compared to a mobile Cortex-A76 such as in the Kirin 990 (which is the best A76 implementation out there), the resulting IPC is 32% better for the Graviton2 in SPECint2006, and 10% better for SPECfp2006. This goes to show what kind of a massive difference the memory subsystem can have on a system that is otherwise similar in terms of the CPU microarchitecture."
Sivar - Friday, March 13, 2020 - link
Generally, doubling cache size does not help performance much unless the cache is far too small already. The memory subsystem is about far more than cache. Memory controller design, clock syncing vs async, trace distances, set associativity, and many other factors come into play.Look at any example in history of cache doubling on a given architecture. You will not see a 30% gain, certainly not an 80% gain.
UnmaskedUnderflow - Friday, March 13, 2020 - link
You're muddling ipc uplift vs power efficiency gain. Yes, on a spec, sub system won't buy you much save in mcf type workloads on raw perf. But you're overlooking the dram effect. On die cache reduces traffic to dram, which for a mobile soc, is exponentially more expensive to access both thermally and energy wise than any horseplay on the soc. The result is real. And even Apple knows this, their massive caches and resultant perf/W are not a mistake. Dram is your mobile enemy for battery life.dotjaz - Sunday, March 15, 2020 - link
Whatever. 66%-94% efficiency jump is just unrealistic no matter how you put it.masimilianzo - Friday, March 13, 2020 - link
In SpecInt2k6 there are tests that are very sensitive to memory latency and others which are very sensitive to memory bandwidth. So you can increase perf a lot.Doubling the L3 cache gives you a lot more perf (Spec data set size is way bigger than 2MB) and also saves on power because you go less often to DDR memory which is crazy high power.
SanX - Friday, March 13, 2020 - link
865 processor sucks versus Apple. The iPhone 11 Pro is whopping 100% faster than S20 on Speedometer2 and JetSctream2 yet Samsung insists in equal prices... HahahahaAlistair - Friday, March 13, 2020 - link
I wish it was equal prices. I don't care about the high end. I'm looking at iPhone 11 vs S20, $700 vs $1000. $300 more, that's crazy.4k HDR - Saturday, March 14, 2020 - link
Wait for 3 month s20 will be cheaper than iPhone 11.Nicon0s - Monday, March 23, 2020 - link
The S20 is a true end to end flagship with a top of the line OLED screen and camera system. It's comparable to the iphone 11 Pro not the plain 11.iPhone 11 basically has the same mediocre screen found in the XR, exactly the same screen, apple mostly only changed the SOC and camera system.
doungmli - Saturday, March 14, 2020 - link
Completely stupid reasoning.1) His tests are developed by webkit (so we imagine safari
advantage)
2) The web tests cannot be compared between different platforms and are used to demonstrate the performance of a browser within the same bone / hardware (They are mainly used to demonstrate the performance of browsers not those of the device and generally developers benefit their browsers For example, in jetstream (develop by webkit) safari is 2x better than firefox, conversely kraken (develop by firefox) firefox is 2x better.
3) To compare, you should use pcmark for example which tests the browsing performance and not the browser
Alistair - Saturday, March 14, 2020 - link
fact is you browse the web with an iphone, it loads faster, so it's fairNicon0s - Sunday, March 15, 2020 - link
It doesn't load faster.Also in general flagship Android phones have better LTE and Wifi performance, that matters more when it comes to webpage loading.
arsjum - Sunday, March 15, 2020 - link
Fact is you can't produce any evidence showing your iPhone loads webpages faster. All I see are benchmarks that don't translate into real life experience.4k HDR - Saturday, March 14, 2020 - link
Equal price for 2 month only after that price start to drops within a year s20 ultra going to cost less than iPhone 11 Pro.Drake H. - Friday, March 13, 2020 - link
These synthetic tests are boring. I'm not going to buy a device to play this. Could you test it on some real-use scene ? Test on heavy games and emulators like Play/Dolphin/DamonPS2.Drake H. - Friday, March 13, 2020 - link
I generally don't like apple products, but I think this time Apple got it right, a 5.8-inch phone(Iphone 11pro) fits perfectly in the hand. These things of almost 7inch are very uncomfortable to use with just one hand. e.eiphonebestgamephone - Friday, March 13, 2020 - link
Those are up on youtube.Psyside - Saturday, March 14, 2020 - link
Games on phones? Facepalm.jpgiphonebestgamephone - Sunday, March 15, 2020 - link
Yes. games on phones. You might want to lookup emulation on android. Huge library.flyingpants265 - Sunday, March 15, 2020 - link
AHHHHHHHHH!!!!!!!N Zaljov - Friday, March 13, 2020 - link
The casually included MT6885Z benchmark results are really sticking out to me. Are you going to review the Redmi K30 or a similar device? If so, I'm really looking forward to it.Following the shambles of S.LSI's previous "adventures" I'm kind of curious as to how SARC's M5 cores end up performing compared to QC's products, but considering the fact that they've scrapped it entirely along with the fact that they're not shipping the Exynos version in their own market (South Korea) along with the struggles of Samsung's Foundry business, I highly doubt that Samsung will be able to be competitive for a very long time...
Anyway: I appreciate your work and hope that you'll continue to deliver excellent content like this in the future.
MrCommunistGen - Friday, March 13, 2020 - link
I also came here to comment on the inclusion of the Dimensity 1000L in the SPEC numbers. Are we getting a full review and/or analysis of a device with this SoC?anonomouse - Friday, March 13, 2020 - link
Also interested, it appears that the supposed 2.6Ghz A77 is performing worse than the 2.4Ghz A77 in SD865, the 2.86Ghz A76 in Kirin 990, and then 2.84Ghz A76 in SD855... which would point to some deficiencies in how Mediatek implemented their memory hierarchy beyond the L1 in their SoC?N Zaljov - Friday, March 13, 2020 - link
Keep in mind that the A77 cores in the Dimensity 1000L are only running with up to 2.2 GHz (it's a slightly downrated version of the Dimensity 1000). If you adjust the SPECSpeed for clocks, the A77 cores should fall in line with the other A77-based SoCs, assuming that there are no bottlenecks that would affect the performance scaling negatively.anonomouse - Monday, March 16, 2020 - link
Ah, forgot about the 1000L vs 1000.Andrei Frumusanu - Friday, March 13, 2020 - link
As mentioned, the D1000L is running at 2.2GHz, not 2.6.Andrei Frumusanu - Friday, March 13, 2020 - link
I'll be doing a piece on that when I'm done with the S20's, yes.Psyside - Saturday, March 14, 2020 - link
Please, please include Flnd X2 Pro in the camera tests, the IMX689 looks amazing so farDragonstongue - Friday, March 13, 2020 - link
IMO I believe for my $ anything considered "flagship" class, should dang well include 3.5mm jack, BT / wireless is "cool" and all that, latency wise however (not to mention connectivity options) 3.5mm jack still has so much more going for it.BT/Wireless has a place, no doubt, but gosh dang, over $1100 phone (above and beyond most full out laptops, which tend to carry audio jacks LOL) then you have to go and buy "respectable" quality earbuds on top of this usually in the $50+ as well.
the world is made of money and we like trashing things every year or 2 "cause that is how we roll"
Oppio (or whatever they called, I typo I know) and some others have very near same spec (not Exynos chip of course) less pricey and many still carry the venerable audio/aux jack not make stupid excuses about not having it either
peevee - Friday, March 13, 2020 - link
TL;DR: in normal mode which will be used by 99% of customers, without the battery-killing "Performance" option enabled, S20 Ultra is about the same as S10+, and even with the option still trails last year iPhones by a whole lot.patel21 - Friday, March 13, 2020 - link
Yes, I just thought the same.I just focused the SD865/60Hz scores, and ignored the 120Hz and P mode.
The reason being even if I or any sane person considering buying S20 series device, its either going to be S20 with its 4000mah battery or the S20+ with 4500mah one.
So when I think of how fast is it from S10+ with the same battery size as S20, the numbers are not interesting at all.
You also need to keep in mind that S20 Ultra is a Huge phone, with lots of body space to cool it down, so the performance numbers might be different from the smaller models, and whatever battery/power the new architecture of SD865 is going to save is more probably going to be wasted by the external modem.
Alistair - Friday, March 13, 2020 - link
Yeah I'm most interested in the S20 normal results, not the + or Ultra. It is already an extra $300 over the iPhone 11, I'm not really interested in the phone that costs double. I'm not interested in the 11 Pro Max etc. either.flyingpants265 - Friday, March 13, 2020 - link
Reviewers love to focus on superficial stuff like specs and benchmarks.Ever notice how flagship phones lag, even on release? The software is EXTREMELY badly written. A lot of these apps should be using maximum 1 megabyte of RAM.. Instead they're loading gigabytes of assets and libraries, makes no sense at all, but makes it much easier on the developers. I wish Terry A. Davis had written a modern smartphone OS. Multi-tasking support should be better, and app-switching should be instantaneous. Even Microsoft Windows can do that.
Oh, and also,
No front speakers = no buy.
No headphone jack = no buy.
And unless it performs miracles (it doesn't), $999 = no buy.
Alistair - Saturday, March 14, 2020 - link
That's why I picked up my first iPhone since the iPhone 4 this week. Cheap, fast, works (yet sadly filled with stupid Apple decisions, but ones I can live with). I want iMessage for the browser, and T9 dialing, basic stuff that Apple refuses to do out of pigheadedness. But it works, battery life (stand by time in particular) is amazing. My less than one year old flagship Android phone, it's Android auto stopped working (the voice assistant), and Google has known about the bugs for 6 months now and hasn't fixed them. Bought the iPhone for Carplay, which is great. Only problem is I use all google services, so I have to append "on Youtube music" to every request, which is annoying.Android and Apple both need to add a default music player setting option, for god's sake.
NickCPC - Saturday, March 14, 2020 - link
I'm a Brit and have had the Exynos 990 S20 Ultra since Monday - really impressed - previous phone being a Pixel 2 XL. Completely agree with the review that the 120Hz mode is a significant step forward - I don't even mind it being a lower resolution. I guess I'm coming forward 3 CPU generations (S835) even though it's less than 2.5yrs since I got the P2XL. Completely expecting the S865 to be superior to the E990 but it's still overall a very good phone. Ironically not had much chance to use the camera yet, but speakers are infinitely better than the P2.4k HDR - Saturday, March 14, 2020 - link
You forget one important test 4K HDR YouTube video with full resolution of the S20 ultra cause S10 Exynos and S10 snapdragon 855 lags if you watch these kind of extreme resolution videos and throttling so fast snapdragon throttle even worst than Exyons.Tlh97 - Sunday, March 15, 2020 - link
Can you add Perf/W efficiency of SD865 GPU in warn condition like in Apple chipsets?Quantumz0d - Sunday, March 15, 2020 - link
Finally some generational improvements like 820 and 835 (efficiency), I think this is a good platform to buy and get it going for a long time since the battery consumption of these planned obsolescence devices will be at-least good enough to keep them going for 3+ years instead of just 2.Coming to SPEC, yeah as usual. Apple's iPhone 7 is defeating this, but one simple question, will the consumer market, Including the supply and demand with respect to the ROI and performance (meaning people's usage patterns and their satisfaction rate) so does these benchmarks translate to the pricing and the user experience ? to put it simply, an iPhone 8 or X as shown in SPEC, will they be able to beat an S10 or S20 experience ? I don't think so at all..
Next up, Samsung S20 series phones are ugly and looks like trash, their camera module is so big and thick and it's an eye sore comparing it to the Note 8,9, S10 this is a huge downgrade, when DJ Koh left Samsung, or since the rumors were coming, I speculated changes but not for worse.
- Lack of SpO2 sensor in Samsung health app now, it's just a pedometer now.
- Removal of 3.5mm jack, what the fuck is this ? This phone is right up at over $1000 without tax and it lacks basic stuff like that for no reason except for greed, their stupid Galaxy Buds do not even have AptX or AptXLL and they use cheap shit Balanced Armatures and demanding over $100 for that ANC, absolute rip off when you see their guts sporting coin cells which could be slot loaded or such but we have another planned obsolescence gadget. AKG is trash, AKG N5005 is shit, they have huge regression in the whole QC, Sound, K3003i was great but this is nothing. Shows in their Galaxy Buds and inbox set.
- Insane price for the base version as well.
- Seeing the camera, they removed that Dual Pixel PDAF and got focusing issues for some nonabinning bullshit.
- Still having a hole despite showing off their Under Screen Camera in 2018, in China.
- Their Knox chip, I wonder Exynos will have bootloader unlock or not this time and how hard is it.
- No stereo speakers which are FF, I know their sound is great on Note10 but it should have been done by now.
- 120Hz for bragging rights, downgraded the display resolution to FHD. It's a shame when their OP7P panel does it at 1440P, they couldn't give the base version at 5K mah ? Zenfone 6 did it last year, LG V60 is doing it this year..
cheetah2k - Monday, March 16, 2020 - link
How does it compare to the exynos 990 S20 Ultra?? This is the question everyone is after!id4andrei - Tuesday, March 17, 2020 - link
Mr. Frumusanu, does Manhattan support Vulkan on Android? They don't advertise Vulkan support on Android but they do support Metal on ios. Roughly the same question for T-Rex.GeorgiaPortus - Saturday, March 21, 2020 - link
https://www.certsofficial.com/">websiteFind [inspiration](https://www.certsofficial.com/) here.
"click here":https://www.certsofficial.com/
ANORTECH - Thursday, March 19, 2020 - link
Why the iPhones are only good in off-screen, its just thermal throttling or is there any other bottleneck? Anyone?Washington Ray - Tuesday, March 24, 2020 - link
What are your Financial needs? We give out loans from a minimum of $2,000.00 to a maximum of $500,000,000.00 with comfortable duration that ranges from 1 to 30 years at a very reduced interest rate of 3%. Do you need a business loan? Do you need a personal loan? Do you want to buy a car? Do you want to refinance? Do you need a mortgage loan? Do you need a huge capital to start off your business proposal or expansion? Have you lost hope and you think there is no way out, and your financial burdens still persists? Please do not hesitate to contact us for possible business co-operation.Our services include the following:
*Refinancing Loans
* Car Loan
*Truck Loans
* Home Loan
* Mortgage Loan
* Debt Consolidation Loan
* Business Loan [secure and unsecured]
* Personal Loan [secure and unsecured]
* Students Loan and so many others.
Sincerely,
Dr. Washington Ray.
(Credit Officer)
Text/Call +1(917)983-0070
([email protected])
Andrew Art - Monday, March 30, 2020 - link
Dear Andrea, could you make an article about Snapdragon 765?Will be great, if you will know about power consumption and power efficiency this chipset and compare this numbers with Snap 865 & 855.
Just try.
iphonebestgamephone - Tuesday, February 9, 2021 - link
Checkout the lg velvet reviewTushar Sharma - Tuesday, April 7, 2020 - link
I wonder why LG and Google have put a full stop on the usage of Snapdragon 865 chipsets.