That would be interesting, to say the least. The power consumption numbers of this part are rather astounding (at least in the Desktop market). I'd like to see how nVidia has managed to make that tolerable (though the lower clock speeds seems to be one answer).
This is a part that is screaming for Optimus, IMO. Couple that with a reasonable CPU (something in the i5 range), and you've got a potent combo.
If CUDA hype is to be believed, then (nearly) everything that you need a quad core CPU for (crunching highly parallel-izeable instructions) is easily accomplished via the GPU (and anything that supports "DirectCompute" at that). I suppose that I never really understood the need for a quad core notebook computer, particularly given what the vast majority of people use their desktops for, and given what marketing says about CUDA...
Updated the text with additional information. The total graphics power (GPU + RAM) of the GTX 480M is 100W, so that's higher than the 75W of previous solutions (GTX 285M). However, there may be better power saving features on the GTX 480M so idle power requirements might be about the same. Given vendors have already started talking about 480M SLI, the power draw is obviously nowhere near as high as the desktop GTX 480 (or even GTX 470)... but then performance is a pretty big step down as well.
I imagine the GTX 480M will offer around half the performance of the desktop GTX 480. The desktop part offers 125% more shader performance, texture fill rate, and 130% more memory bandwidth. Half of the GTX 470 might be a closer estimate.
I'm pretty sure this is a die-harvested GF100, nVidia wouldn't make a GF104 chip with identical specs to the GTX 465, which we know is a GF100 chip. My expectation is that the GF104 will be a 256-shader part (half of a uncut Fermi). That part makes much more sense to use in the Mobile space, I suspect they used the GTX 465 because they couldn't wait for the GF104 to be out, ready, and binned for the mobile market before taking advantage of the enthusiast market by launching a product.
NVIDIA has confirmed with me that this is a die harvested GF100. The future products aren't even out for the desktop market, so it's highly unlikely they would release first in the mobile sector. I expect we'll see mobile versions of the other GF10x chips down the road. As for the 100W... well, companies put Core i7 desktop parts into large notebooks, so why not a 100W GPU? You can fit two 285M chips in an 18.4" notebook as well, so a single 100W GPU isn't that much worse. Obviously, these aren't even remotely targeting the portable market but are after people that don't mind large, heavy, but fast notebooks.
Yeah, at 100W, it just doesn't make sense as a mobile product anymore. Remember when 100W was ridiculous for a desktop GPU or CPU? Now they call that "mobile" class thermal specs. Eventually a notebook can only get so heavy and so thick before you might as well just haul a mATX system around.
Sorry to call you an idiot but you are an idiot. If you need a laptop that delivers 10h - buy something with integrated graphics and ultra low power CPU with ultra low performance.
I do need a powerful laptop that I could easily carry around and that it still has the power of a desktop PC and I don't care about battery.
Yes there is need and market for these monsters and you are not forced to buy it. For you - buy an iPad :)
Also, for some users mobility means "close the lid and put the laptop in sleep mode until I go to the conference room". Certainly there are the users for which mobility means "a mobile phone is too heavy, and I want to work all the way from USA to Singapore, why isn't my laptop working 20 hours from a charge?" Fermi for laptops will fit in a niche inside the very small niche of performance gaming laptops - however, there will be some people willing to buy it
you know, I am of the same mind... I am sick to Fing death of all those space heater 300 watt desktop GPUs with a fan louder then a turbine and an electic bill that makes me weep... also it BURNS MY FEET if I keep them too close to the case's vent! (I am talking literally here, it hurts!)
Anyone who wants great performance, that's who. It's no more than what notebooks are already shipping with. In fact it's LESS than what notebooks have already shipped with. It's a kick ass part, and I love that we're getting it. I'll probably order a system with it if I can find one that's okay (the ongoing Asus/Dell weirdness issues with various systems makes finding one that's good iffy, but the M17x-R2 has been a good system so far, and they might put it in that).
If this is only half a GTX 480performance wise, wouldn't a desktop HD 5770 use the same amount of power as this card with similar performance for only $150 dollars? I think that these high end mobile GPUs are relativity expensive so using a stock desktop GPU instead of a $300? GTX 480M would make a killer laptop for much cheaper. You could use the same types of chassis/cooling as the GTX 480M needs.
On the other hand the Mobility HD 5870 is just an underclocked HD 5770 designed to use only 60W. Using the same binned RV840 chips used in the Mobility 5870 and going all the way to 100W, ATI could make a mobile GPU that is an overclocked desktop HD 5770 and easily meet nvidia's performance levels.
Nvidia can't really win in the mobile GPU space with these blazing hot chips.
I never really did understand the "super performance" GPUs in laptops when they are such a significantly small segment of the market that you can't make any reasonable amount of money off of them. Even in desktops, far more money is spent in GPUs that cost under $200 than those that are $400+.
We need to ditch this pure performance metric and start working in two-dimensional statistics: performance per watt for laptops and performance per dollar for desktops.
The only small minded people here are the ones like you.
You people dont seem to understand that not everyone has YOUR specific needs and wants. This GPU might not be good for you or even most people, but it is for others.
Just looking at the gigaflops provided shows that the Radeon Mobility 5870 is vastly superior. So is the Radeon Mobility 5850 and the Radeon Mobility 5830.
But GFLOPS is the single most useless spec around I think, particularly for GPUs. Pure, theoretical performance is one thing; what you can actually get in practice is another. Look at desktop GTX 480 vs. HD 5870 and you'll see the same GFLOPS "deficit", but NVIDIA is consistently faster. Now power on the other hand....
Google:" DIY Vidock" bandwidth is tight but soon to improve with PCI Express 2.0 works like a charm,
never again to buy a heavy expansive notebook that its gpu turns out dead in less than two years (nvidia) and from a manufacturer (gateway) that does not have spare parts to support its own products.
With a mobile part as powerful as Nvidia is claiming this to be it seems like it might make a good fit for either an ITX mobo or something along the lines of book sized HTPC system that could double as a 1080P gaming system. I mean some of those systems use sodimms so why not use other parts designed for notebooks. Just an idea I'm posing.
Another high-power, high-heat part that makes it even less advisable to have a "laptop" anywhere near one's family jewels.
So, if it can't go more than about 10 feet between charges, and you can't actually have it in your lap without risking burns or fried eggs, how is it not a slightly portable desktop with ultra-small form factor?
"Also worth noting is that the mobile DX11 at present consists of a single top-end GPU, whereas AMD has top to bottom DX11 support."
It sounds like you're saying that ATI doesn't have a mobile Crossfire solution for the top end right now, but you can get two of the 5870 (mobile) in the Alienware m17x (and probably in other systems, I just don't care to look).
Yeah... all I meant is that NVIDIA just has the mobile 480M. ATI has mobile HD 5870, 5850, 5830, 5770, 5750, 5730, 5650, 5470, 5450, and 5430. (Note that HD 5145 and 5165 are just rebadged DX10.1 solutions.) So ATI has what you need from the top to bottom with DX11 support, and NVIDIA focuses just on the top right now. Of course, once you get below the 5830, the actual utility of DX11 becomes limited. The 5650 runs medium details + DX11 "okay" at 1366x768; anything more and you're generally better off dropping to DX10/DX9 modes.
I linked to the GTX 280M vs. HD 5870 review of the G73Jh, where the 5870 is about 10-15% faster than GTX 280M. 285M would make the gap even smaller, and it looks like GTX 480M should be about 30-50% faster than 285M. The HD 5850 is going to be about 12% slower than the 5870, which means it would be more like a GTX 280M and thus clearly slower than the 480M.
Well since we are talking mobile I was kind of more interested in the watt to rendering power ratio. I mean I wouldn't mind taking a 20% fps hit if its going to use half the power. Though I would want all the features the same or better. DX11 hardware codec acceleration etc.
In terms of power, the 5870 uses a maximum of 50W I think (not sure if that figure includes the RAM or not), so unless the 480M is twice as fast ATI obviously wins there. The 5850 uses up to 39W (22% less power) and delivers 11% less performance, so perf/watt goes up. However, keep in mind that idle power draw on the 5830, 5850, and 5870 should be the same, as they'll all run at the lowest possible clock and voltage when idle.
Despite the horrid 100W TDP, I have to commend Nvidia for actually putting a desktop GTX480 chip in the mobile GTX480 (even if it is only about half as powerful). They have rebadged the G92 as GTX280 and GTX380 for far too long. It is good to see a return to more reasonable naming for mobile parts.
So how big is the charger brick going to be? Total system draw with one of these could probably top 150w, which is the size of the giant 360 power brick (not that that is a good comparison, since I think they have left the size the same since it was 250w or whatever it started with)
But yeah, that would be a lot of power in a notebook. However, in my experience when you run SLI notebooks on battery power, the GPUs are forced into a lower performance, lower power state. The battery just isn't capable of putting out that much power I don't think. In fact, I once tested an Alienware notebook with an ATI X1800 GPU, and the system could draw so much power that the battery life could actually drain while playing games plugged in! LOL. (Or maybe that was just a horrible design flaw? Probably....)
Do they make games for PCs anymore? Anyone else getting frustrated that publishers are forgetting the PC as a gaming platform with all of this great hardware coming out?
I've read that the GTX480M Can bitstream the HD audio codecs Dolby TrueHD and DTS-MA over HDMI rather than just send as Multi Channel LPCM as the desktop 480/470 variant's do.
Can anybody confirm and if so what are the chances of the upcoming GTX465 being able to do the same?
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
deputc26 - Tuesday, May 25, 2010 - link
352 shaders?? Please tell me this not a die-harvested gtx400 ($10 says it is).retnuh - Tuesday, May 25, 2010 - link
Ignore the man behind the curtain......erple2 - Tuesday, May 25, 2010 - link
That would be interesting, to say the least. The power consumption numbers of this part are rather astounding (at least in the Desktop market). I'd like to see how nVidia has managed to make that tolerable (though the lower clock speeds seems to be one answer).This is a part that is screaming for Optimus, IMO. Couple that with a reasonable CPU (something in the i5 range), and you've got a potent combo.
If CUDA hype is to be believed, then (nearly) everything that you need a quad core CPU for (crunching highly parallel-izeable instructions) is easily accomplished via the GPU (and anything that supports "DirectCompute" at that). I suppose that I never really understood the need for a quad core notebook computer, particularly given what the vast majority of people use their desktops for, and given what marketing says about CUDA...
JarredWalton - Tuesday, May 25, 2010 - link
Updated the text with additional information. The total graphics power (GPU + RAM) of the GTX 480M is 100W, so that's higher than the 75W of previous solutions (GTX 285M). However, there may be better power saving features on the GTX 480M so idle power requirements might be about the same. Given vendors have already started talking about 480M SLI, the power draw is obviously nowhere near as high as the desktop GTX 480 (or even GTX 470)... but then performance is a pretty big step down as well.I imagine the GTX 480M will offer around half the performance of the desktop GTX 480. The desktop part offers 125% more shader performance, texture fill rate, and 130% more memory bandwidth. Half of the GTX 470 might be a closer estimate.
descendency - Tuesday, May 25, 2010 - link
Mobile??? How far can you move in 10 minutes?mianmian - Tuesday, May 25, 2010 - link
Nvidia re-defined the word: move from one plug to another while in sleeping mode.dagamer34 - Tuesday, May 25, 2010 - link
Batteries in the laptops that have a GTX480M class card serve more as an uninterruptible power supply than anything else.Griswold - Wednesday, May 26, 2010 - link
To the trash container.Drag0nFire - Wednesday, May 26, 2010 - link
I can't even imagine the heat these things must put out in a laptop chassis...taltamir - Tuesday, May 25, 2010 - link
acording to xbitlabs it would be a GF104 based card, not GF100 (thus, not a die harvested GTX480... the GTX465 is a die harvested GTX480 though)Anyways, 100W is a bad joke... who in their right mind would put a 100W GPU in a laptop?
ekv - Tuesday, May 25, 2010 - link
Isn't Apple running their processors at 100C?LtGoonRush - Tuesday, May 25, 2010 - link
I'm pretty sure this is a die-harvested GF100, nVidia wouldn't make a GF104 chip with identical specs to the GTX 465, which we know is a GF100 chip. My expectation is that the GF104 will be a 256-shader part (half of a uncut Fermi). That part makes much more sense to use in the Mobile space, I suspect they used the GTX 465 because they couldn't wait for the GF104 to be out, ready, and binned for the mobile market before taking advantage of the enthusiast market by launching a product.JarredWalton - Tuesday, May 25, 2010 - link
NVIDIA has confirmed with me that this is a die harvested GF100. The future products aren't even out for the desktop market, so it's highly unlikely they would release first in the mobile sector. I expect we'll see mobile versions of the other GF10x chips down the road. As for the 100W... well, companies put Core i7 desktop parts into large notebooks, so why not a 100W GPU? You can fit two 285M chips in an 18.4" notebook as well, so a single 100W GPU isn't that much worse. Obviously, these aren't even remotely targeting the portable market but are after people that don't mind large, heavy, but fast notebooks.MonkeyPaw - Tuesday, May 25, 2010 - link
Yeah, at 100W, it just doesn't make sense as a mobile product anymore. Remember when 100W was ridiculous for a desktop GPU or CPU? Now they call that "mobile" class thermal specs. Eventually a notebook can only get so heavy and so thick before you might as well just haul a mATX system around.Zingam - Wednesday, May 26, 2010 - link
Sorry to call you an idiot but you are an idiot. If you need a laptop that delivers 10h - buy something with integrated graphics and ultra low power CPU with ultra low performance.I do need a powerful laptop that I could easily carry around and that it still has the power of a desktop PC and I don't care about battery.
Yes there is need and market for these monsters and you are not forced to buy it. For you - buy an iPad :)
Calin - Monday, May 31, 2010 - link
Also, for some users mobility means "close the lid and put the laptop in sleep mode until I go to the conference room".Certainly there are the users for which mobility means "a mobile phone is too heavy, and I want to work all the way from USA to Singapore, why isn't my laptop working 20 hours from a charge?"
Fermi for laptops will fit in a niche inside the very small niche of performance gaming laptops - however, there will be some people willing to buy it
Ken g6 - Tuesday, May 25, 2010 - link
Actually, 100W sounds perfect...for a desktop GPU! When can I get one of those?taltamir - Wednesday, May 26, 2010 - link
you know, I am of the same mind... I am sick to Fing death of all those space heater 300 watt desktop GPUs with a fan louder then a turbine and an electic bill that makes me weep... also it BURNS MY FEET if I keep them too close to the case's vent! (I am talking literally here, it hurts!)Griswold - Wednesday, May 26, 2010 - link
At 100W for the desktop, you get something better - from the other team.Wolfpup - Sunday, May 30, 2010 - link
Anyone who wants great performance, that's who. It's no more than what notebooks are already shipping with. In fact it's LESS than what notebooks have already shipped with. It's a kick ass part, and I love that we're getting it. I'll probably order a system with it if I can find one that's okay (the ongoing Asus/Dell weirdness issues with various systems makes finding one that's good iffy, but the M17x-R2 has been a good system so far, and they might put it in that).Zink - Tuesday, May 25, 2010 - link
If this is only half a GTX 480performance wise, wouldn't a desktop HD 5770 use the same amount of power as this card with similar performance for only $150 dollars? I think that these high end mobile GPUs are relativity expensive so using a stock desktop GPU instead of a $300? GTX 480M would make a killer laptop for much cheaper. You could use the same types of chassis/cooling as the GTX 480M needs.On the other hand the Mobility HD 5870 is just an underclocked HD 5770 designed to use only 60W. Using the same binned RV840 chips used in the Mobility 5870 and going all the way to 100W, ATI could make a mobile GPU that is an overclocked desktop HD 5770 and easily meet nvidia's performance levels.
Nvidia can't really win in the mobile GPU space with these blazing hot chips.
dagamer34 - Tuesday, May 25, 2010 - link
I never really did understand the "super performance" GPUs in laptops when they are such a significantly small segment of the market that you can't make any reasonable amount of money off of them. Even in desktops, far more money is spent in GPUs that cost under $200 than those that are $400+.We need to ditch this pure performance metric and start working in two-dimensional statistics: performance per watt for laptops and performance per dollar for desktops.
GullLars - Wednesday, May 26, 2010 - link
+1, single dimensioned thinking is for small minded people ;)B3an - Wednesday, May 26, 2010 - link
The only small minded people here are the ones like you.You people dont seem to understand that not everyone has YOUR specific needs and wants.
This GPU might not be good for you or even most people, but it is for others.
André - Wednesday, May 26, 2010 - link
Well, this is where it gets funny.Just looking at the gigaflops provided shows that the Radeon Mobility 5870 is vastly superior. So is the Radeon Mobility 5850 and the Radeon Mobility 5830.
JarredWalton - Wednesday, May 26, 2010 - link
But GFLOPS is the single most useless spec around I think, particularly for GPUs. Pure, theoretical performance is one thing; what you can actually get in practice is another. Look at desktop GTX 480 vs. HD 5870 and you'll see the same GFLOPS "deficit", but NVIDIA is consistently faster. Now power on the other hand....sh_kamalh - Wednesday, May 26, 2010 - link
External GPU is the way to go.I wish for a 14'' laptop with Core i5 and an external GPU on the same level of desktop 57xx+
Griswold - Wednesday, May 26, 2010 - link
ATI tried to establish that a couple years ago. Nobody wanted it.ik9000 - Monday, June 7, 2010 - link
well, shuttle seem to think differently: http://tv.hexus.net/show/2010/06/Computex_2010_Sou... notebook boosted by external desktop graphics card being showed off at computex 2010.york50 - Wednesday, May 26, 2010 - link
Google:" DIY Vidock"bandwidth is tight but soon to improve with PCI Express 2.0
works like a charm,
never again to buy a heavy expansive notebook that its gpu turns out dead in less than two years (nvidia) and from a manufacturer (gateway) that does not have spare parts to support its own products.
BAH
spidey81 - Wednesday, May 26, 2010 - link
With a mobile part as powerful as Nvidia is claiming this to be it seems like it might make a good fit for either an ITX mobo or something along the lines of book sized HTPC system that could double as a 1080P gaming system. I mean some of those systems use sodimms so why not use other parts designed for notebooks. Just an idea I'm posing.Omophorus - Wednesday, May 26, 2010 - link
Another high-power, high-heat part that makes it even less advisable to have a "laptop" anywhere near one's family jewels.So, if it can't go more than about 10 feet between charges, and you can't actually have it in your lap without risking burns or fried eggs, how is it not a slightly portable desktop with ultra-small form factor?
Pessimism - Wednesday, May 26, 2010 - link
I just wish they would stop constantly changing the MXM mobile graphics card format and mandate it in all laptops.theeldest - Wednesday, May 26, 2010 - link
"Also worth noting is that the mobile DX11 at present consists of a single top-end GPU, whereas AMD has top to bottom DX11 support."It sounds like you're saying that ATI doesn't have a mobile Crossfire solution for the top end right now, but you can get two of the 5870 (mobile) in the Alienware m17x (and probably in other systems, I just don't care to look).
Unless I'm reading this wrong?
JarredWalton - Wednesday, May 26, 2010 - link
Yeah... all I meant is that NVIDIA just has the mobile 480M. ATI has mobile HD 5870, 5850, 5830, 5770, 5750, 5730, 5650, 5470, 5450, and 5430. (Note that HD 5145 and 5165 are just rebadged DX10.1 solutions.) So ATI has what you need from the top to bottom with DX11 support, and NVIDIA focuses just on the top right now. Of course, once you get below the 5830, the actual utility of DX11 becomes limited. The 5650 runs medium details + DX11 "okay" at 1366x768; anything more and you're generally better off dropping to DX10/DX9 modes.synaesthetic - Thursday, May 27, 2010 - link
The 5650 doesn't really even run DX11 "okay" at 1366x768... my lappy has one and I find myself disabling those features more often than not.JarredWalton - Thursday, May 27, 2010 - link
Note that I said "medium details and DX11"... if you try to run high/ultra quality, forget it. LOLSunSamurai - Wednesday, May 26, 2010 - link
To this?http://www.notebookcheck.net/ATI-Mobility-Radeon-H...
JarredWalton - Wednesday, May 26, 2010 - link
I linked to the GTX 280M vs. HD 5870 review of the G73Jh, where the 5870 is about 10-15% faster than GTX 280M. 285M would make the gap even smaller, and it looks like GTX 480M should be about 30-50% faster than 285M. The HD 5850 is going to be about 12% slower than the 5870, which means it would be more like a GTX 280M and thus clearly slower than the 480M.SunSamurai - Wednesday, May 26, 2010 - link
Well since we are talking mobile I was kind of more interested in the watt to rendering power ratio. I mean I wouldn't mind taking a 20% fps hit if its going to use half the power. Though I would want all the features the same or better. DX11 hardware codec acceleration etc.JarredWalton - Wednesday, May 26, 2010 - link
In terms of power, the 5870 uses a maximum of 50W I think (not sure if that figure includes the RAM or not), so unless the 480M is twice as fast ATI obviously wins there. The 5850 uses up to 39W (22% less power) and delivers 11% less performance, so perf/watt goes up. However, keep in mind that idle power draw on the 5830, 5850, and 5870 should be the same, as they'll all run at the lowest possible clock and voltage when idle.teko - Wednesday, May 26, 2010 - link
First laptop with 2 power chargers? :PSlaimus - Wednesday, May 26, 2010 - link
Despite the horrid 100W TDP, I have to commend Nvidia for actually putting a desktop GTX480 chip in the mobile GTX480 (even if it is only about half as powerful). They have rebadged the G92 as GTX280 and GTX380 for far too long. It is good to see a return to more reasonable naming for mobile parts.smartalco - Wednesday, May 26, 2010 - link
So how big is the charger brick going to be? Total system draw with one of these could probably top 150w, which is the size of the giant 360 power brick (not that that is a good comparison, since I think they have left the size the same since it was 250w or whatever it started with)Zink - Thursday, May 27, 2010 - link
55W - i7 920XM200W - GTX480M SLI
An SLI system could push 250W. That would be a power brick. Even at 150W you would be lucky to get 45min out of the battery gaming.
JarredWalton - Thursday, May 27, 2010 - link
There are 250W power bricks I'm pretty sure... certainly 230W and I think 240W. See the ASUS W90Vp:http://images.anandtech.com/galleries/452/asus-w90...
But yeah, that would be a lot of power in a notebook. However, in my experience when you run SLI notebooks on battery power, the GPUs are forced into a lower performance, lower power state. The battery just isn't capable of putting out that much power I don't think. In fact, I once tested an Alienware notebook with an ATI X1800 GPU, and the system could draw so much power that the battery life could actually drain while playing games plugged in! LOL. (Or maybe that was just a horrible design flaw? Probably....)
Chadder007 - Thursday, May 27, 2010 - link
Do they make games for PCs anymore? Anyone else getting frustrated that publishers are forgetting the PC as a gaming platform with all of this great hardware coming out?mark0409mr01 - Thursday, May 27, 2010 - link
I've read that the GTX480M Can bitstream the HD audio codecs Dolby TrueHD and DTS-MA over HDMI rather than just send as Multi Channel LPCM as the desktop 480/470 variant's do.Can anybody confirm and if so what are the chances of the upcoming GTX465 being able to do the same?
kallogan - Monday, May 31, 2010 - link
ROFL Nvidia. That's all.