I'm going to guess no since the Exynos SoC with AMD's GPU IP is likely to fabricated using Samsung's 5LPE process node, which are generally inferior to TSMC's N7P process node when power efficiency is concerned, especially at higher frequencies.
Wasn't it like 1-2% better? Also, they'll probably use the next iteration of that like 4LPP or 5LPP which likely be just slightly better than TSMC's N7P.
Andrei mentioned that Samsung's 5LPE process node is roughly on par with TSMC's N7P at lower performance levels. But at higher performance levels, Samsung's 5LPE process node consumed 25% more power in comparison to TSMC's N7P process node. https://www.anandtech.com/show/16463/snapdragon-88...
just how how fucking stupid AMD's ex CEO Hector Ruiz was when he sold IMAGEON and the mobile developers for beans right before the explosion of mobile smart phones.
Hindsight can be deceptive, because we focus mainly on the opportunity he missed and not all the factors weighing on the decision. I'm not trying to defend it, but I do wonder why they did it, and what other options they might've had to get the money they presumably needed.
Back in the 90's we developed the Tinma project. It was a complete PC on a single chip. Then we scrapped the whole thing because obviously nobody will ever buy that.
I remember in the early 2000's, when wearable computing was a sort of fringe community of weirdo gadget people. Then, within a few years, smartphones happened and the idea of carrying around a computer with you suddenly became completely normalized.
I suspect it will not be branded as RDNA 3, but 'SS Exynos Next-Gen Blah-Blah-Blah' (by AMD / Radeon?) ___ gotta let the marketing dept do their thing with cross-branding
Because ray tracing is probably part of a high quality AR SW/HW solution. And why do you want AR on your phone? Because the way you get to a decent smart glasses solution is by starting with an adequate solution on a larger, easier to deal with, piece of hardware.
Or Android can, you know, let Apple continue to increase its multi-year lead over everyone else in wearables.
This is a good point. Game devs will want similar features on all devices.
However, the non-Samsung Android market is still too large for developers to ignore. So, they're going to have to put at least some effort into devices without VRS or RT for some time yet to come.
Even though the non-Samsung Android market is very large, Samsung is still a power brand in that ecosystem. If they adopt ray tracing in their mobile phones, other Android brands would probably want to follow suit. Hopefully Qualcomm will deliver something similar soon.
As for Apple, I wouldn't be surprised if they also adopt real time ray tracing in an upcoming Ax SoC as they use IMG's IP for their in-house GPU designs.
Maybe Samsung intends this to enable mobile gaming and console combined. My kids play their switch mostly with it in the HDMI cradle and hooked to a TV. My Razer 2 phone already sells a phone case that can attach/detach game paddles similar to the Nitendo Switch. VRS and FSR would help lighten the load with Ray-Tracing and hopefully make it look good also on a big screen TV.
Raytracing isn't just displaying pretty graphics. You can use it for other utilitarian purposes as well. The first thing that jumps to mind is all those fancy filters on your camera. You can use raytracing on the depth sensor's data output to determine locations and movement.
Interesting idea. Although it normally works on the basis of traversing a bounding-volume hierarchy, which takes some work to build (as does postprocessing depth sensor camera output). So, if you were hoping that would save power vs. doing image-based comparisons to spot movement, I say probably not.
One thing you could use it for is audio environment simulation. Overkill for casual listening, but if we're just talking about other things it can do...
I suppose when you are a hammer everything looks like a nail. Samsung has stated that the primary purpose of this SOC is Windows 10 on ARM laptops with one being produced "this summer":
Samsung didn't say that; "unnamed sources" did. FWIW I think it's entirely plausible, but it's not as if it's confirmed.
I think, BTW, that the writers here are probably too optimistic with their prediction that the early Radeon Exynos parts will be RDNA3. RDNA2 seems far more likely based on the timing.
RDNA 2 is very much an 2021/22 product. Next gen APUs are expected to have RDNA 2 (have only been Vega until now) and there are still RDNA 2 discrete products coming out. RDNA 3 is likely some way away.
The thing is, it is unlikely to be ready. Polaris was in AMD GPUs before consoles. Vega came first as consumer AMD GPUs, then AMD APUs and last to Mac GPUs. RDNA came first to AMD GPUs then to Mac GPUs. RDNA2 came to CPUs before consoles and it is rumoured to come to APUs (Zen 3+).
Samsung will incorporate the GPU into the S22 that might come as early as January. RDNA3 is not expected that soon. I wouldn't exclude a mix between RDNA2 and RDNA3 though.
A counterpoint would be the Intel Vega chip that actually was more Polaris than Vega. It shows that AMD might play fast and loose with the naming of semi-custom products. They might for instance call an almost RDNA3 product RDNA2 for recognition as console chips, and to be able to call the next generation RDNA3 without having to do a full generation.
Nvidia Tegra chips have one to two year old graphics architectures. An example is the Nintendo switch that has the same architecture as GeForce 700-series.
To conclude, RDNA2 is completely feasible for Samsung's next chip. It might share more details with the rumoured AMD APUs than with the desktop CPUs though.
> Nvidia Tegra chips have one to two year old graphics architectures.
Speaking specifically of the X1, it actually had some features before the bigger Nvidia GPUs got them. IIRC, it was very much part of the then-current Maxwell generation, not lagging.
> the Nintendo switch that has the same architecture as GeForce 700-series.
No, 900-series. I think there was one Maxwell chip released as a 700-series card, but nearly all of them were 900's.
Yea the GTX 750 Ti and 750 were weird cards. Super fast for their power consumption at the time though -- I recommended the TI to someone who I didn't think had researched his power supply well enough and got infinite kudos for the rest of the school term.
> RDNA 2 as a graphics IP is a 2020 matter and Samsung’s next-gen Exynos SoC > coming out at the end of 2021 for 2022 ... very much indicate this is an RDNA 3 IP
I think the way AMD likes to operate is to get their new graphics IP fully baked and debugged in dGPUs, first. Then propagate it to other products. Same thing for their CPU cores, and it explains why the APUs are always lagging.
In Samsung's case, they'll want tested and debugged IP, before they start porting it to their node and integrating it with their SoC. So, I think some lag is to be expected. This does seem a bit more lag than I'd expect, but maybe the timetables just didn't line up for RDNA 3 to be ready in time for Samsung.
Mobile SoC vendors move at a much much faster pace than the likes of AMD who have a lot more validation work to do - tape out to mass production happens in 8 months.
Regarding the latest RDNA2/3 point: wouldn't surprise me if it's just RDNA2 or something like 3.5. After all, AMD yet to release mobile APUs with RDNA instead of Vega.
Now that x86 AMD Radeon 6xxxM-RDNA2 is rolling, the next step is to cut-down the 'M' graphics engines-VCN/ASICs for APUs. For the most part, AMD 'rolls' this way, a step out of cadence.
Vega is the odd step-child whereby on-going performance/efficiency gains were just too much to pass up (against the competition). They maintained their 'AMD-vantage'
The big kicker in all of this 3D V-Cache and graphic chiplet design. No telling what Dr Su and her teams have up their sleeves ... AMD ARM big-LITTLE SOCs ? Make it so!
I don't think most people are going to care about ray tracing on a phone - BUT, this could be big for future standalone VR headsets. I so badly want something other than the Quest 2 as a real option in the US, if someone latches on to this and we can start getting better, more immersive lighting in VR environments without being tethered to a PC...that's a big deal.
I do think that this program will be beneficial for both parties. I hope that companies that have online roulette games get together and form a online roulette strategy partnership soon. I say that because just like this program is beneficial, this partnership will be too. <a href="https://www.fiverr.com/faizan_5478/convert-psd-to-... Web Development</a>
The only problem with this ,is samsung chip design. It always never power efficient compared to snapdragon. Even 2100 feels like a design fix instead of generation upgrade , if they can not implement a good power efficiency this means nothing
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
49 Comments
Back to Article
yeeeeman - Tuesday, June 1, 2021 - link
hopefully it will be better than the 6000m gpus when it comes to efficiency.i highly doubt it will match adreno from qualcomm.
Dakhil - Tuesday, June 1, 2021 - link
I'm going to guess no since the Exynos SoC with AMD's GPU IP is likely to fabricated using Samsung's 5LPE process node, which are generally inferior to TSMC's N7P process node when power efficiency is concerned, especially at higher frequencies.flashmozzg - Tuesday, June 1, 2021 - link
Wasn't it like 1-2% better?Also, they'll probably use the next iteration of that like 4LPP or 5LPP which likely be just slightly better than TSMC's N7P.
Dakhil - Tuesday, June 1, 2021 - link
Andrei mentioned that Samsung's 5LPE process node is roughly on par with TSMC's N7P at lower performance levels. But at higher performance levels, Samsung's 5LPE process node consumed 25% more power in comparison to TSMC's N7P process node. https://www.anandtech.com/show/16463/snapdragon-88...alumine - Tuesday, June 1, 2021 - link
Ahh that would be a little ironic given the Adreno guys were ex-ATI...Adreno descended from the Imageon and is also an anagram of Radeon :)
tamalero - Friday, June 4, 2021 - link
just how how fucking stupid AMD's ex CEO Hector Ruiz was when he sold IMAGEON and the mobile developers for beans right before the explosion of mobile smart phones.mode_13h - Sunday, June 6, 2021 - link
Hindsight can be deceptive, because we focus mainly on the opportunity he missed and not all the factors weighing on the decision. I'm not trying to defend it, but I do wonder why they did it, and what other options they might've had to get the money they presumably needed.JKflipflop98 - Sunday, June 6, 2021 - link
Back in the 90's we developed the Tinma project. It was a complete PC on a single chip. Then we scrapped the whole thing because obviously nobody will ever buy that.mode_13h - Monday, June 7, 2021 - link
I remember in the early 2000's, when wearable computing was a sort of fringe community of weirdo gadget people. Then, within a few years, smartphones happened and the idea of carrying around a computer with you suddenly became completely normalized.Smell This - Tuesday, June 1, 2021 - link
I suspect it will not be branded as RDNA 3, but 'SS Exynos Next-Gen Blah-Blah-Blah' (by AMD / Radeon?) ___ gotta let the marketing dept do their thing with cross-branding
The Hardcard - Tuesday, June 1, 2021 - link
The question of RDNA2 vs. RDNA3 would be based on the IP and features implemented, not the timeframe. Vega is also a 2021 matter.Tomatotech - Tuesday, June 1, 2021 - link
Why would I want ray-tracing on a phone?Flunk - Tuesday, June 1, 2021 - link
marketing.iphonebestgamephone - Tuesday, June 1, 2021 - link
Thats just one of the reasons.s.yu - Wednesday, June 9, 2021 - link
Well, if there's no hope of it matching even RTX2060...name99 - Tuesday, June 1, 2021 - link
Because ray tracing is probably part of a high quality AR SW/HW solution.And why do you want AR on your phone? Because the way you get to a decent smart glasses solution is by starting with an adequate solution on a larger, easier to deal with, piece of hardware.
Or Android can, you know, let Apple continue to increase its multi-year lead over everyone else in wearables.
taigebu - Tuesday, June 1, 2021 - link
Less work for game devs like no need to bake lighting.mode_13h - Tuesday, June 1, 2021 - link
This is a good point. Game devs will want similar features on all devices.However, the non-Samsung Android market is still too large for developers to ignore. So, they're going to have to put at least some effort into devices without VRS or RT for some time yet to come.
taigebu - Tuesday, June 1, 2021 - link
Even though the non-Samsung Android market is very large, Samsung is still a power brand in that ecosystem. If they adopt ray tracing in their mobile phones, other Android brands would probably want to follow suit. Hopefully Qualcomm will deliver something similar soon.As for Apple, I wouldn't be surprised if they also adopt real time ray tracing in an upcoming Ax SoC as they use IMG's IP for their in-house GPU designs.
Freeb!rd - Tuesday, June 1, 2021 - link
Maybe Samsung intends this to enable mobile gaming and console combined. My kids play their switch mostly with it in the HDMI cradle and hooked to a TV. My Razer 2 phone already sells a phone case that can attach/detach game paddles similar to the Nitendo Switch. VRS and FSR would help lighten the load with Ray-Tracing and hopefully make it look good also on a big screen TV.iphonebestgamephone - Tuesday, June 1, 2021 - link
For the same reason as on pc? Have you never seen raytracing?serendip - Sunday, June 6, 2021 - link
On Windows on ARM chips.mode_13h - Sunday, June 6, 2021 - link
Right, for laptops. Conforming with DX12 Ultimate would be a good selling-point.JKflipflop98 - Sunday, June 6, 2021 - link
Raytracing isn't just displaying pretty graphics. You can use it for other utilitarian purposes as well. The first thing that jumps to mind is all those fancy filters on your camera. You can use raytracing on the depth sensor's data output to determine locations and movement.mode_13h - Monday, June 7, 2021 - link
Interesting idea. Although it normally works on the basis of traversing a bounding-volume hierarchy, which takes some work to build (as does postprocessing depth sensor camera output). So, if you were hoping that would save power vs. doing image-based comparisons to spot movement, I say probably not.One thing you could use it for is audio environment simulation. Overkill for casual listening, but if we're just talking about other things it can do...
igor velky - Tuesday, June 1, 2021 - link
there will be NO RDNA3 in samsung soc announced this year.fishingbait15 - Tuesday, June 1, 2021 - link
I suppose when you are a hammer everything looks like a nail. Samsung has stated that the primary purpose of this SOC is Windows 10 on ARM laptops with one being produced "this summer":https://www.sammobile.com/news/exynos-2200-soc-wit...
MediaTek and Nvidia are working on a competing SOC that will likely turn up in Chromebooks in 2022:
https://www.notebookcheck.net/Nvidia-partnering-wi...
SarahKerrigan - Tuesday, June 1, 2021 - link
Samsung didn't say that; "unnamed sources" did. FWIW I think it's entirely plausible, but it's not as if it's confirmed.I think, BTW, that the writers here are probably too optimistic with their prediction that the early Radeon Exynos parts will be RDNA3. RDNA2 seems far more likely based on the timing.
mode_13h - Tuesday, June 1, 2021 - link
Agreed.mode_13h - Tuesday, June 1, 2021 - link
Yeah, it'd be really interesting to see AMD's own IP used to compete against them, in the chromebook market.Would get REALLY interesting, if/when AMD decides to get (back) into the ARM game.
dontlistentome - Tuesday, June 1, 2021 - link
I suspect Gen 1 will have great performance, but (comparitively) terrible efficiency. It's Gen 2 you need to watch out for.Fataliity - Tuesday, June 1, 2021 - link
They've been working on it for 3-4 years. I'm willing to bet it took this long because they want it to be competitive.ET - Tuesday, June 1, 2021 - link
RDNA 2 is very much an 2021/22 product. Next gen APUs are expected to have RDNA 2 (have only been Vega until now) and there are still RDNA 2 discrete products coming out. RDNA 3 is likely some way away.nandnandnand - Tuesday, June 1, 2021 - link
No reason that Samsung can't get early access to RDNA 3 IP for a custom implementation.Rudde - Wednesday, June 2, 2021 - link
The thing is, it is unlikely to be ready. Polaris was in AMD GPUs before consoles. Vega came first as consumer AMD GPUs, then AMD APUs and last to Mac GPUs. RDNA came first to AMD GPUs then to Mac GPUs. RDNA2 came to CPUs before consoles and it is rumoured to come to APUs (Zen 3+).Samsung will incorporate the GPU into the S22 that might come as early as January. RDNA3 is not expected that soon. I wouldn't exclude a mix between RDNA2 and RDNA3 though.
A counterpoint would be the Intel Vega chip that actually was more Polaris than Vega. It shows that AMD might play fast and loose with the naming of semi-custom products. They might for instance call an almost RDNA3 product RDNA2 for recognition as console chips, and to be able to call the next generation RDNA3 without having to do a full generation.
Nvidia Tegra chips have one to two year old graphics architectures. An example is the Nintendo switch that has the same architecture as GeForce 700-series.
To conclude, RDNA2 is completely feasible for Samsung's next chip. It might share more details with the rumoured AMD APUs than with the desktop CPUs though.
mode_13h - Friday, June 4, 2021 - link
> Nvidia Tegra chips have one to two year old graphics architectures.Speaking specifically of the X1, it actually had some features before the bigger Nvidia GPUs got them. IIRC, it was very much part of the then-current Maxwell generation, not lagging.
> the Nintendo switch that has the same architecture as GeForce 700-series.
No, 900-series. I think there was one Maxwell chip released as a 700-series card, but nearly all of them were 900's.
lmcd - Tuesday, September 28, 2021 - link
Yea the GTX 750 Ti and 750 were weird cards. Super fast for their power consumption at the time though -- I recommended the TI to someone who I didn't think had researched his power supply well enough and got infinite kudos for the rest of the school term.wrkingclass_hero - Tuesday, June 1, 2021 - link
This is likely the reason AMD's mobile APUs still use Vega.nandnandnand - Tuesday, June 1, 2021 - link
Not at all.mode_13h - Tuesday, June 1, 2021 - link
> RDNA 2 as a graphics IP is a 2020 matter and Samsung’s next-gen Exynos SoC> coming out at the end of 2021 for 2022 ... very much indicate this is an RDNA 3 IP
I think the way AMD likes to operate is to get their new graphics IP fully baked and debugged in dGPUs, first. Then propagate it to other products. Same thing for their CPU cores, and it explains why the APUs are always lagging.
In Samsung's case, they'll want tested and debugged IP, before they start porting it to their node and integrating it with their SoC. So, I think some lag is to be expected. This does seem a bit more lag than I'd expect, but maybe the timetables just didn't line up for RDNA 3 to be ready in time for Samsung.
Andrei Frumusanu - Tuesday, June 1, 2021 - link
Mobile SoC vendors move at a much much faster pace than the likes of AMD who have a lot more validation work to do - tape out to mass production happens in 8 months.flashmozzg - Tuesday, June 1, 2021 - link
>I think the way AMD likes to operate is to get their new graphics IP fully baked and debugged in dGPUs, first.I think it's mostly due to cadence mismatch and collab/sync difficulties between CPU and GPU teams more than anything.
flashmozzg - Tuesday, June 1, 2021 - link
Regarding the latest RDNA2/3 point: wouldn't surprise me if it's just RDNA2 or something like 3.5. After all, AMD yet to release mobile APUs with RDNA instead of Vega.Smell This - Thursday, June 3, 2021 - link
I suspect it is that 'cadence mismatch' thang.
Now that x86 AMD Radeon 6xxxM-RDNA2 is rolling, the next step is to cut-down the 'M' graphics engines-VCN/ASICs for APUs. For the most part, AMD 'rolls' this way, a step out of cadence.
Vega is the odd step-child whereby on-going performance/efficiency gains were just too much to pass up (against the competition). They maintained their 'AMD-vantage'
The big kicker in all of this 3D V-Cache and graphic chiplet design. No telling what Dr Su and her teams have up their sleeves ... AMD ARM big-LITTLE SOCs ? Make it so!
jefmes - Friday, June 4, 2021 - link
I don't think most people are going to care about ray tracing on a phone - BUT, this could be big for future standalone VR headsets. I so badly want something other than the Quest 2 as a real option in the US, if someone latches on to this and we can start getting better, more immersive lighting in VR environments without being tethered to a PC...that's a big deal.nandnandnand - Friday, June 4, 2021 - link
Samsung phones can be used as docked desktops.Let's see what happens when ray-tracing is combined with FSR on one of these phones.
alex43 - Saturday, June 5, 2021 - link
I do think that this program will be beneficial for both parties. I hope that companies that have online roulette games get together and form a online roulette strategy partnership soon. I say that because just like this program is beneficial, this partnership will be too. <a href="https://www.fiverr.com/faizan_5478/convert-psd-to-... Web Development</a>mode_13h - Sunday, June 6, 2021 - link
spammerMidioni2 - Monday, June 28, 2021 - link
The only problem with this ,is samsung chip design. It always never power efficient compared to snapdragon. Even 2100 feels like a design fix instead of generation upgrade , if they can not implement a good power efficiency this means nothing