SFF.Network NVIDIA Set to Announce Pascal GTX GPU's Today

We've seen dribs and drabs of NVIDIA's upcoming Pascal refresh for their consumer graphics cards - from leaks of reference shrouds, to rumored (if not highly questionable) benchmarks. But the most recent slew of data has enough substance to suggest that it's the real deal, and the timing of the leak just before NVIDIA's announcement of a livestream event later today all but confirms that we're about to see the next generation of consumer flagships.

Eagle-eyed folks at videocardz.com caught some benchmarks published without identifiers indicating their GPU of origin (though driver names did spill the beans), and from that they've been able to construct a semi-complete specifications table that compares the rumored GTX 1080 and 1070 against the recently-released Tesla P100...

Read more here.

 

[PlayfulPhoenix]

After seeing so many GTX 980 ti special editions release in the last few months, I can't help but think a lot of people will be royally pissed if they were sold a totally obsolete card at such a high price that was released only months before a brand new model (if the 1070 is really superior to it). If the 980 ti is at least competitive with the 1080 in non VR gaming, then I think they can at least tell themselves that it's still a good card.

I mean, this happens every time, right? Especially when something has been on the market for longer than is typical (As Maxwell has), you should pretty much expect a big leap for the next generation of silicon.

I generally don't empathize with people complaining that companies are releasing upgrades that are too substantial or too quick (talk about a first world problem). And anyone even passively paying attention would have seen the many signs that updates to the 970 and 980 were coming

It looks like AMD have responded to NVidia's recent 1070/1080 reveal by bringing forward the launch of Vega10 (with HBM2 memory)from early 2017 to October 2016.

If AMD wasn't anticipating the 1070/1080, I'm worried. Moving a launch date forward 4+ months, when it's less than a year out, is not something you do when you think you're competitive.

It is believed that Vega10 will compete with the 1070/1080 market sector, whereas the upcoming Polaris will compete in the low to mid range market.

The Polaris architecture may only compete at the mid to low end of the market, but the 14nm could mean even better fps per watt performance than what NVidia have just announced based on 16nm, can't wait for the first reviews and comparisons to start coming in once both sides have products on the market.

AMD's own language suggests that Polaris is going to be the bottom segment of the discrete graphics market, so there's no reason to expect otherwise. Whether or not that means they'll release a part that's in the vicinity of the 1070 performance-wise is the real question - them suggesting "affordable VR" makes me excited, but we can only wait and see for now.

The efficiency of the parts will be interesting, as well. One thing AMD could do that would be really exciting is to have as many of the cards as possible be powered via PCI alone - the segment and price point make that feasible, and the demand's there for sure.

 

[EdZ]

AMD's own language suggests that Polaris is going to be the bottom segment of the discrete graphics market, so there's no reason to expect otherwise. Whether or not that means they'll release a part that's in the vicinity of the 1070 performance-wise is the real question - them suggesting "affordable VR" makes me excited, but we can only wait and see for now.

To be supremely pessimistic: Oculus' stated minimum spec for AMD cards is "AMD R9 290 equivalent or greater", so that's not a massive performance requirement for Polaris 10 to fulfil and still meet AMD's claim.

 

[PlayfulPhoenix]

To be supremely pessimistic: Oculus' stated minimum spec for AMD cards is "AMD R9 290 equivalent or greater", so that's not a massive performance requirement for Polaris 10 to fulfil and still AMD's claim.

Totally true. We're swimming in ambiguity right now, on both sides.

Fortunately, we shouldn't have to wait too long before we have a firm grasp on Pascal and Polaris alike.

 

[IntoxicatedPuma]

I mean, this happens every time, right? Especially when something has been on the market for longer than is typical (As Maxwell has), you should pretty much expect a big leap for the next generation of silicon.

I generally don't empathize with people complaining that companies are releasing upgrades that are too substantial or too quick (talk about a first world problem). And anyone even passively paying attention would have seen the many signs that updates to the 970 and 980 were coming

I don't think that's a very fair statement to make. Most people don't follow the industry heavily. I know 4 people where I work (at a video game developer) who would have bought a GTX 970 in the last 3 months if I had not begged them to wait. I also know two people who bought GTX 980 ti's in the last few months. All of these people are enthusiastic gamers, they just don't follow the industry that closely.

The best comparison I can make is, Nvidia is selling these cards like Ferrari or Porsche sells an outgoing car model. Ferrari is going to release a new model so they start selling special edition versions of the outgoing one that are high performance and have a high price. Except in those cases, they're collectors cars which will probably have a high value for a long time to come and may be worth more later on. In addition, new cars are announced long before they go on sale, so buyers usually know what they're getting.

I will say that EVGA has a good policy regarding this. The "Step-Up" is really considerate towards buyers. I specifically was thinking of a card like the GTX 980 ti "VR Edition" which was released on Feb 11 of this year. Buyers of that card would have been screwed a bit since the 1080 will no doubt destroy it in VR performance. I see that the Step-Up cutoff date was set to Feb 11 so seems they're aware that buyers of that card are probably going to want to upgrade to the 1080.

I hope other companies have policies like this, EVGA is the only one that I know of but maybe others do. I agree that some people complain that performance improvements don't come often, but it's not fair to say that people should be given the shaft for buying high end cards only to watch them be totally outclassed within a few months of the product being released.

 

[iFreilicht]

If AMD wasn't anticipating the 1070/1080, I'm worried. Moving a launch date forward 4+ months, when it's less than a year out, is not something you do when you think you're competitive.

Is it not, though? To me it just feels like it reinforces my belief in the R9 Nano and Fury not being competitive as soon as the nVidia 10 series are released. So, if AMD suspects they won't be able to cut the price sufficiently, it makes sense to release their new chips earlier.
Of course it seems strange that they actually can do that. A lot of stuff happens during 4 Months, so if they can announce that release now, they're either heavily cutting functionality somewhere or have made a huge development leap and wanted to wait with announcing it.

 

[jØrd]

Its AMD, they probably just shit-canned driver development to get it out the door quicker

 

[GuilleAcoustic]

Its AMD, they probably just shit-canned driver development to get it out the door quicker

Lol, it's sad but true.

 

[K888D]

Oooooo, these new Zotac cards like nice, shiny:

http://www.overclock3d.net/articles...1080_pgf_gpu_and_a_new_overclocking_utility/1

 

[EdZ]

Videocardz have some PCB shots of the reference GTX 1080. I made a quick blink comparison with the GTX 980 PCB. It;s clear Nvidia see no reason to re-invent the wheel.

Oooooo, these new Zotac cards like nice, shiny:

http://www.overclock3d.net/articles...1080_pgf_gpu_and_a_new_overclocking_utility/1

Very Tron. A shame edge-illuminated plastic never looks that nice and evenly illuminated in person.

 

[PlayfulPhoenix]

I don't think that's a very fair statement to make. Most people don't follow the industry heavily. I know 4 people where I work (at a video game developer) who would have bought a GTX 970 in the last 3 months if I had not begged them to wait. I also know two people who bought GTX 980 ti's in the last few months. All of these people are enthusiastic gamers, they just don't follow the industry that closely.

The best comparison I can make is, Nvidia is selling these cards like Ferrari or Porsche sells an outgoing car model. Ferrari is going to release a new model so they start selling special edition versions of the outgoing one that are high performance and have a high price. Except in those cases, they're collectors cars which will probably have a high value for a long time to come and may be worth more later on. In addition, new cars are announced long before they go on sale, so buyers usually know what they're getting.

I mean, I get what you're saying, it's just that I don't think you can really blame NVIDIA or AMD for this dynamic. They aren't obligated in any way to disclose what they're working on or when it's released, and I don't think anyone would argue as such. Companies are free to sell what they have for as much as they want, and share information as they choose, so long as it's honest. The same goes for people.

Consequently, if it's anyone's "fault" that someone bought a 970 or 980 (or even a Titan X) a few weeks ago, well, the only other entity in that exchange is the buyer. The buyer made the choice to purchase something, and it's the buyer's responsibility to inform their own purchases. If they didn't know an update was coming even after leaks and rumors and everything - all of which a quick Google search would have shown - then that's on them. Just as if I buy a crappy monitor because I didn't spend five minutes reading professional reviews, then that's on me, as well. The information was freely and readily available, but the buyer didn't perform even basic due diligence before spending hundreds and hundreds of dollars.

The other thing, too, is that it isn't as if buying a card just before a refresh is bad in every way, either - if those folks bought those cards three months ago, then hey, they would have had an upgraded experience for months now. That isn't worth nothing! Even if it isn't as ideal as waiting in the long run, they couldn't even argue that they didn't get something they otherwise wouldn't have. So even if I feel bad for them that they didn't make the best purchasing decision they could, I'm not going to feel like they were cheated, or place the responsibility for that on anyone else.

To me it just feels like it reinforces my belief in the R9 Nano and Fury not being competitive as soon as the nVidia 10 series are released. So, if AMD suspects they won't be able to cut the price sufficiently, it makes sense to release their new chips earlier.

Oh, for sure. I'm sure I'll get a lot of heat for saying this, but AMD's current lineup is one of the most compromised in recent memory - it's a hodgepodge of re-branded old cards and "flagships" that have poorly chosen compromises, from 4GB of VRAM to essentially mandatory use of AIO's. And I'd bet that most folks at AMD would agree with that, too, behind closed doors, because they've been saying all along that the future is Polaris and HBM2 and 14nm and everything. The problem is, people are buying cards right now, and presently on the AMD side we're in the ugly transitional period.

AMD has paid dearly for that - 3 out of 4 graphics cards sold today are NVIDIA. That's not because 75% of the market is in the tank for NVIDIA. That's because NVIDIA's products are better for most people currently.

Of course it seems strange that they actually can do that. A lot of stuff happens during 4 Months, so if they can announce that release now, they're either heavily cutting functionality somewhere or have made a huge development leap and wanted to wait with announcing it.

Or they wanted to spread out releases across their multi-year timeline (desirable given how hard generational leaps are becoming), thought they'd have more time to respond to NVIDIA, and realized that they didn't. But even then, they're having to ramp up production in half the time, compared to a few days ago. Assuming that this rumor is true, AMD's got their work cut out for them.

 

[Phuncz]

Lol, it's sad but true.

For me it isn't. I got the R9 290X reference the day they came out and since the start it has been nothing but smooth sailing. Maybe not every game in the world, but all the games I played worked wonderfully. It all turned to shit when I installed Windows 10 and apparently I'm not the only one that has issues on that OS with any type or brand of GPU. No problem if I use Windowed Borderless display mode in games and/or close RainMeter. I think the number of times I had a crash on Windows 7 not related to a bad game patch can be counted on a couple of fingers.

The only problem I've had with AMD's drivers in the last few years is that they switched their DVI to HDMI logic and somehow my TV with HDMI connected to the DVI on my GPU makes it believe it has audio capabilities, limiting it to RGB 4:2:0. I fabricated a custom monitor driver so the audio/extended capabilities are hidden and now it works at RGB 4:4:4 like it should.

 

[GuilleAcoustic]

Last AMD card that I had was an HD5870 and I only had problems with it sadly.

 

[Ceros_X]

Personally, I believe that the jon of companies is to make products consumers want and to make money for shareholders (for publically traded companies). If someone is smart enough to install and play a videogame, then they are smart enough to do a little googling before buying a graphics card - especially if they are replacing computer components on their own, etc.

If someone bought a 980Ti three months before the next Gen cards launched, they didn't buy a bad card. The release of new cards doesn't turn the 980 into crap. They just bought a good card and now there is better. Next time do your research. Their might be better cards out in a month or two, but the 980 or 970 should still be able to play any games on the market right now.

 

[jØrd]

For me it isn't.

To be clear i was being somewhat tongue in cheek cynical. That said i was also referring to my experience w/ their Linux driver. On arch i have to run a patched x.org & the catalyst-test packages from the AUR. AMD do do a better job w/ their OSS driver and their new unified GPU driver strategy (OSS kernel driver, closed and OSS user space packages) but ive heard this story about how AMD's next driver thing on Linux will be great many times before now. Needless to say im jumping ship to Nvidia for the build im currently part sourcing. The OSS Nvidia drivers are hot garbage but their closed source drivers are, realistically, the only choice for gaming on Linux. Unless you want to get into VR, then your SOL regardless

I fabricated a custom monitor driver

Totally smooth sailing right there

 

[Phuncz]

That said i was also referring to my experience w/ their Linux driver.

I'm dreading that too, unless I hardware passthrough the dGPU to Windows and I use the Intel CPU's iGPU in Linux, for non-gaming. But indeed, AMD's proper Linux support has been "arriving" for years now.

ThatTotally smooth sailing right there

It doesn't require any programming knowledge, just edit a file with an app and use another app to write the INF file. I found a guide online, I'll look it up when I get back from work.

EDIT: The links I had were out of date, but it's a general issue also for Nvidia:
http://www.tested.com/tech/pcs/457468-troubleshooting-dvi-and-edid-issues/
http://www.avsforum.com/forum/166-l...official-4-4-4-chroma-subsampling-thread.html

 

[K888D]

Some 3DMARK scores have been leaked for the 1080, at stock and OC clock speeds. They are not confirmed though.

The 1080 is showing around 65% more performance than the current 980 at stock speeds. Overclocking to 2.1Ghz results in a further 24% performance.

http://www.overclock3d.net/articles/gpu_displays/gtx_1080_overclocked_3dmark_performance_leaked/1

 

[PlayfulPhoenix]

It doesn't require any programming knowledge, just edit a file with an app and use another app to write the INF file.

This reminds me of the time I had to do very similar work to get SLI GTX 285's to function on a Hackintosh. I couldn't find anyone that had done it before at the time, and had to author my own kexts to get it to work. Spent 15+ hours over a long weekend, and eventually somehow managed to get my tri-monitor rig functioning.

I still vividly remember when I rebooted for what felt like the millionth time, and I saw the familiar grey screen only on my center monitor, once again facing defeat... and then - as the OS flashed blue before loading the desktop - all three monitors lit up in a sea of aquamarine, before revealing my wallpaper and windows spread across all three screens.

To this day, it's one of my most favorite moments with respect to computers and building as a hobby.

Some 3DMARK scores have been leaked for the 1080, at stock and OC clock speeds. They are not confirmed though.

The 1080 is showing around 65% more performance than the current 980 at stock speeds. Overclocking to 2.1Ghz results in a further 24% performance.

Drip, drip, drip...

Dashes of salt are still warranted, but these numbers do make sense when put next to the limited gaming benchmarks that NVIDIA themselves have shared. Their website shows 1.6-1.8x improvement (when comparing stock 980 to stock 1080) for "traditional", non-VR gaming. And, doing the quick math:

LEAKED (UNCONFIRMED) 3DMARK SCORES

980: 13148
980 Ti: 17042
1080: 21828
1080@2.1: 26456

1080 vs 980: (21828-13148) / 13148 = 1.66x improvement

That's not really >= 980 SLI, but it's pretty close - I'd put a 980 SLI setup at around 1.7-1.8x a sole 980, when accounting for the inefficiencies of SLI.

More interesting, and more exciting, is that OC benchmark, given that NVIDIA got to 2.1Ghz on their reference cooler in their demonstration. We won't know how reliable you can hit that until we see lots of reviews, and this is doubly the case for cards that don't use the Founders Edition shroud. But NVIDIA was conservative if anything when they discussed Maxwell's OC-ability, so there's certainly the potential that most folks could get to 2.1 with relatively little effort. Plus, more performance-oriented coolers will only lift the thermal ceiling higher (though, as with Maxwell, that might not even be what limits overclocking in the end).

1080@2.1 vs 980: (26456-13148)/13148 = 2.01x improvement

Now that's beating a 980 SLI any day of the week.

 

[EdZ]

I remember having to do an EDID-flash with an 1920x1200 LCD (an LG, IIRC) that expected a Reduced Blanking DVI timing mode but advertised regular DVI timings. That was a pain.

The GTX 1080 is certainly becoming a more tempting upgrade over the 980ti. I'll be waiting for proper benchmarking to see what sort of improvement is evident for VR though.

 

[Phuncz]

While Nvidia is adament about comparing the 1080 vs the 980, price-wise it's going to be the 1080 vs 980Ti which to me matters most. Simply because I hope most buy a label or a brand, but a product and as such take performance and price into consideration. Although I see the merit in comparing a previous generation card to it's successor, that previous card has been available for 25% less for about a year.

The 980 also came out at a much higher price than currently, so for all the people with a GTX 980Ti, it might be more interesting to wait out for that price drop or to wait for a possible GTX 1080Ti.

A GTX 980 vs GTX 1080 with 65% speed increase sounds like an awesome upgrade.
A GTX 980Ti vs GTX 1080 with 20% speed increase sounds much less enticing.
Especially since a second-hand GTX 980Ti seems to go at about 2/3rds the GTX 1080's price.

 

[K888D]

Some Polaris specs have leaked (unconfirmed). As previously thought, they will not be competing with the 1070 and 1080.

The interesting part though is that there is rumoured to be a card with a TDP of 50W offering 2.5TFLOPs of compute performance. This should make a nice PCIe Powered card, great for SFF PC's.

http://www.overclock3d.net/articles/gpu_displays/amd_polaris_10_and_polaris_11_specifications_leak/1