GP104 Strikes Again
It’s only been three weeks since NVIDIA unveiled the GeForce GTX 1080 and GTX 1070 graphics cards at a live streaming event in Austin, TX. But it feels like those two GPUs, one of which hasn't even been reviewed until today, have already drastically shifted the landscape of graphics, VR and PC gaming.
Half of the “new GPU” stories are told, with AMD due to follow up soon with Polaris, but it was clear to anyone watching the enthusiast segment with a hint of history that a line was drawn in the sand that day. There is THEN, and there is NOW. Today’s detailed review of the GeForce GTX 1070 completes NVIDIA’s first wave of NOW products, following closely behind the GeForce GTX 1080.
Interestingly, and in a move that is very uncharacteristic of NVIDIA, detailed specifications of the GeForce GTX 1070 were released on GeForce.com well before today’s reviews. With information on the CUDA core count, clock speeds, and memory bandwidth it was possible to get a solid sense of where the GTX 1070 performed; and I imagine that many of you already did the napkin math to figure that out. There is no more guessing though - reviews and testing are all done, and I think you'll find that the GTX 1070 is as exciting, if not more so, than the GTX 1080 due to the performance and pricing combination that it provides.
Let’s dive in.
Subject: Graphics Cards | May 29, 2016 - 05:46 PM | Scott Michaud
Tagged: msi, GTX 1080, sea hawk, gaming x, armor, Aero, nvidia
Beyond the Founders Edition, MSI has prepared six SKUs of the GTX 1080. These consists of four variants, two of which have an overclocked counterpart to make up the remaining two products. The product stack seems quite interesting, with a steady progression of user needs, but we'll need to wait for price and availability to know for sure.
We'll start at the bottom with the MSI GeForce GTX 1080 AERO 8G and MSI GeForce GTX 1080 AERO 8G OC. These are your typical blower designs that pull air in from within the case, and exhausts it out the back after collecting a bunch of heat from the GPU. It will work, it should be one of the cheapest options for this card, and it will keep the GTX 1080's heat outside of the case. It has a little silver accent on it, too. The non-overclocked version is the standard 1607 MHz / 1733 MHz that NVIDIA advertises, and the OC SKU is a little higher: 1632 MHz / 1771 MHz.
Next up the product stack are the MSI GeForce GTX 1080 ARMOR 8G and MSI GeForce GTX 1080 ARMOR 8G OC versions. This uses MSI's aftermarket, two-fan cooler that should provide much lower temperatures than AERO, but they exhaust back into the case. Personally? I don't really care about that. The only other thing that heats up in my case, to any concerning level at least, is my CPU, and I recently switched that to a closed-loop water cooler anyway. MSI added an extra, six-pin power connector to these cards (totaling 8-pin + 6-pin + slot power = up-to 300W, versus 8-pin + slot power's 225W). The non-overclocked version is NVIDIA's base 1607 MHz / 1733 MHz, but OC brings that up to 1657 MHz / 1797 MHz.
Speaking of closed-loop water coolers... The MSI GeForce GTX 1080 SEA HAWK takes the AERO design, which we mentioned earlier, and puts a Corsair self-contained water cooler inside it, too. Only one SKU of this is available, clocked at 1708 MHz base and 1847 MHz boost, but it should support overclocking fairly easily. That said, unlike other options that add a bonus six-pin connector, the SEA HAWK has just one, eight-pin connector. Good enough for the Founders Edition, but other SKUs (including three of the other cards in this post) suggest that there's a reason to up the power ceiling.
We now get to MSI's top, air-cooled SKU: the MSI GeForce GTX 1080 GAMING X 8G. This one has their new TWIN FROZR VI, which they claim spins quieter and has fans that drag more air to spin slower than previous models. It, as you would assume from reading about ARMOR 8G, has an extra, six-pin power connector to provide more overclocking headroom. It has three modes: Silent, which clocks the card to the standard 1607 MHz / 1733 MHz levels; Gaming, which significantly raises that to 1683 MHz / 1822 MHz; and OC, which bumps that slightly further to 1708 MHz / 1847 MHz.
Currently, no pricing and availability for any of these.
Subject: Graphics Cards | May 28, 2016 - 05:00 PM | Scott Michaud
Tagged: asus, ROG, strix, GTX 1080, nvidia
The Founders Edition versions of the GTX 1080 went on sale yesterday, but we're beginning to see the third-party variants being announced. In this case, the ASUS ROG Strix is a three-fan design that uses their DirectCU III heatsink. More interestingly, ASUS decided to increase the amount of wattage that this card can accept by adding an extra, six-pin PCIe power connector (totaling 8-pin + 6-pin). A Founders Edition card only requires a single, eight-pin connection over the 75W provided by the PCIe slot itself. This provides an extra 75W of play room for the ROG Strix card, raising the maximum power from 225W to 300W.
Some of this power will be used for its on-card, RGB LED lighting, but I doubt that it was the reason for the extra 75W of headroom. The lights follow the edges of the card, acting like hats and bow-ties to the three fans. (Yes, you will never unsee that now.) The shroud is also modular, and ASUS provides the data for enthusiasts to 3D print their own modifications (albeit their warranty doesn't cover damage caused by this level of customization).
As for the actual performance, the card naturally comes with an overclock out of the box. The default “Gaming Mode” has a 1759 MHz base clock with an 1898 MHz boost. You can flip this into “OC Mode” for a slight, two-digit increase to 1784 MHz base and 1936 MHz boost. It is significantly higher than the Founders Edition, though, which has a base clock of 1607 MHz that boosts to 1733 MHz. The extra power will likely help manual overclocks, but it will come down to “silicon lottery” whether your specific chip was abnormally less influenced by manufacturing defects. We also don't know yet whether the Pascal architecture, and the 16nm process it relies upon, has any physical limits that will increasingly resist overclocks past a certain frequency.
Pricing and availability is not yet announced.
First, Some Background
NVIDIA's Rumored GP102
When GP100 was announced, Josh and I were discussing, internally, how it would make sense in the gaming industry. Recently, an article on WCCFTech cited anonymous sources, which should always be taken with a dash of salt, that claimed NVIDIA was planning a second architecture, GP102, between GP104 and GP100. As I was writing this editorial about it, relating it to our own speculation about the physics of Pascal, VideoCardz claims to have been contacted by the developers of AIDA64, seemingly on-the-record, also citing a GP102 design.
I will retell chunks of the rumor, but also add my opinion to it.
In the last few generations, each architecture had a flagship chip that was released in both gaming and professional SKUs. Neither audience had access to a chip that was larger than the other's largest of that generation. Clock rates and disabled portions varied by specific product, with gaming usually getting the more aggressive performance for slightly better benchmarks. Fermi had GF100/GF110, Kepler had GK110/GK210, and Maxwell had GM200. Each of these were available in Tesla, Quadro, and GeForce cards, especially Titans.
Maxwell was interesting, though. NVIDIA was unable to leave 28nm, which Kepler launched on, so they created a second architecture at that node. To increase performance without having access to more feature density, you need to make your designs bigger, more optimized, or more simple. GM200 was giant and optimized, but, to get the performance levels it achieved, also needed to be more simple. Something needed to go, and double-precision (FP64) performance was the big omission. NVIDIA was upfront about it at the Titan X launch, and told their GPU compute customers to keep purchasing Kepler if they valued FP64.
Subject: Graphics Cards | May 24, 2016 - 06:36 PM | Scott Michaud
Tagged: nvidia, graphics drivers
Yesterday, NVIDIA has released WHQL-certified drivers to align with the release of Overwatch. This version, 368.22, is the first public release of the 367 branch. Pascal is not listed in the documentation as a supported product, so it's unclear whether this will be the launch driver for it. The GTX 1080 comes out on Friday, but two drivers in a week would not be unprecedented for NVIDIA.
While NVIDIA has not communicated this too well, 368.22 will not install on Windows Vista. If you are still using that operating system, then you will not be able to upgrade your graphics drivers past 365.19. 367-branch (and later) drivers will required Windows 7 and up.
Before I continue, I should note that I've experienced so issues getting these drivers to install through GeForce Experience. Long story short, it took two attempts (with a clean install each time) to end up with a successful boot into 368.22. I didn't try the standalone installer that you can download from NVIDIA's website. If the second attempt using GeForce Experience failed, then I would have. That said, after I installed it, it seemed to work out well for me with my GTX 670.
While NVIDIA is a bit behind on documentation, the driver also rolls in other fixes. There were some GPU compute developers who had crashes and other failures in certain OpenCL and CUDA applications, which are now compatible with 368.22. I've also noticed that my taskbar hasn't been sliding around on its own anymore, but I've only been using the driver for a handful of hours.
You can get GeForce 368.22 drivers from GeForce Experience, but you might want to download the standalone installer (or skip a version or two if everything works fine).
Subject: General Tech | May 19, 2016 - 12:08 PM | Ryan Shrout
Tagged: video, radeon, polaris 11, polaris 10, Polaris, podcast, pascal, nvidia, GTX 1080, gtx 1070, gtx, geforce, arm, amd, 10nm
PC Perspective Podcast #400 - 05/19/2016
Join us this week as we discuss the GTX 1080 performance and features, official specifications of the GTX 1070, new Polaris specification rumors, ARM's 10nm chip test and more!
The URL for the podcast is: http://pcper.com/podcast - Share with your friends!
- iTunes - Subscribe to the podcast directly through the Store (audio only)
- Google Play - Subscribe to our audio podcast directly through Google Play!
- RSS - Subscribe through your regular RSS reader (audio only)
- MP3 - Direct download link to the MP3 file
Hosts: Ryan Shrout, Jeremy Hellstrom, Josh Walrath, and Allyn Malventano
Subject: Graphics Cards | May 18, 2016 - 12:49 PM | Josh Walrath
Tagged: nvidia, pascal, gtx 1070, 1070, gtx, GTX 1080, 16nm FF+, TSMC, Founder's Edition
Several weeks ago when NVIDIA announced the new GTX 1000 series of products, we were given a quick glimpse of the GTX 1070. This upper-midrange card is to carry a $379 price tag in retail form while the "Founder's Edition" will hit the $449 mark. Today NVIDIA released the full specifications of this card on their website.
The interest of the GTX 1070 is incredibly great because of the potential performance of this card vs. the previous generation. Price is also a big consideration here as it is far easier to raise $370 than it is to make the jump to GTX 1080 and shell out $599 once non-Founder's Edition cards are released. The GTX 1070 has all of the same features as the GTX 1080, but it takes a hit when it comes to clockspeed and shader units.
The GTX 1070 is a Pascal based part that is fabricated on TSMC's 16nm FF+ node. It shares the same overall transistor count of the GTX 1080, but it is partially disabled. The GTX 1070 contains 1920 CUDA cores as compared to the 2560 cores of the 1080. Essentially one full GPC is disabled to reach that number. The clockspeeds take a hit as well compared to the full GTX 1080. The base clock for the 1070 is still an impressive 1506 MHz and boost reaches 1683 MHz. This combination of shader counts and clockspeed makes this probably a little bit faster than the older GTX 980 ti. The rated TDP for the card is 150 watts with a single 8 pin PCI-E power connector. This means that there should be some decent headroom when it comes to overclocking this card. Due to binning and yields, we may not see 2+ GHz overclocks with these cards, especially if NVIDIA cut down the power delivery system as compared to the GTX 1080. Time will tell on that one.
The memory technology that NVIDIA is using for this card is not the cutting edge GDDR5x or HBM, but rather the tried and true GDDR5. 8 GB of this memory sits on a 256 bit bus, but it is running at a very, very fast 8 gbps. This gives overall bandwidth in the 256 GB/sec region. When we combine this figure with the memory compression techniques implemented with the Pascal architecture we can see that the GTX 1070 will not be bandwidth starved. We have no information if this generation of products will mirror what we saw with the previous generation GTX 970 in terms of disabled memory controllers and the 3.5 GB/500 MB memory split due to that unique memory subsystem.
Beyond those things, the GTX 1070 is identical to the GTX 1080 in terms of DirectX features, display specifications, decoding support, double bandwidth SLI, etc. There is an obvious amount of excitement for this card considering its potential performance and price point. These supposedly will be available in the Founder's Edition release on June 10 for the $449 MSRP. I know many people are considering using these cards in SLI to deliver performance for half the price of last year's GTX 980ti. From all indications, these cards will be a signficant upgrade for anyone using GTX 970s in SLI. With the greater access to monitors that hit 4K as well as Surround Gaming, this could be a solid purchase for anyone looking to step up their game in these scenarios.
Subject: Graphics Cards | May 17, 2016 - 06:22 PM | Jeremy Hellstrom
Tagged: nvidia, pascal, video, GTX 1080, gtx, GP104, geforce, founders edition
Yes that's right, if you felt Ryan and Al somehow missed something in our review of the new GTX 1080 or you felt the obvious pro-Matrox bios was showing here are the other reviews you can pick and choose from. Start off with [H]ard|OCP who also tested Ashes of the Singularity and Doom as well as the old favourite Battlefield 4. Doom really showed itself off as a next generation game, its Nightmare mode scoffing at any GPU with less than 5GB of VRAM available and pushing the single 1080 hard. Read on to see how the competition stacked up ... or wait for the 1440 to come out some time in the future.
"NVIDIA's next generation video card is here, the GeForce GTX 1080 Founders Edition video card based on the new Pascal architecture will be explored. We will compare it against the GeForce GTX 980 Ti and Radeon R9 Fury X in many games to find out what it is capable of."
Here are some more Graphics Card articles from around the web:
- In the lab: Nvidia's GeForce GTX 1080 graphics card @ The Tech Report
- FCAT GeForce GTX 1080 Framepacing @ Guru of 3D
- NVIDIA GeForce GTX 1080 Review: A Look At 4K & Ultra-wide Gaming @ Techgage
- NVIDIA GeForce GTX 1080 Review - The Advent of Pascal @HiTech Legion
- NVIDIA GeForce GTX 1080 Founders Edition Review @ OCC
- NVIDIA GeForce GTX 1080 Founders Edition Review @ Neoseeker
- Nvidia GTX 1080 @ Kitguru
- NVIDIA GeForce GTX 1080 8 GB @ techPowerUp
- The NVIDIA GeForce GTX 1080 Review @ Hardware Canucks
A new architecture with GP104
Table of Contents
- Asynchronous compute discussion
- Is only 2-Way SLI supported?
- Overclocking over 2.0 GHz
- Dissecting the Founders Edition
- Benchmarks begin
- VR Testing
- Impressive power efficiency
- Performance per dollar discussion
- Ansel screenshot tool
The summer of change for GPUs has begun with today’s review of the GeForce GTX 1080. NVIDIA has endured leaks, speculation and criticism for months now, with enthusiasts calling out NVIDIA for not including HBM technology or for not having asynchronous compute capability. Last week NVIDIA’s CEO Jen-Hsun Huang went on stage and officially announced the GTX 1080 and GTX 1070 graphics cards with a healthy amount of information about their supposed performance and price points. Issues around cost and what exactly a Founders Edition is aside, the event was well received and clearly showed a performance and efficiency improvement that we were not expecting.
The question is, does the actual product live up to the hype? Can NVIDIA overcome some users’ negative view of the Founders Edition to create a product message that will get the wide range of PC gamers looking for an upgrade path an option they’ll take?
I’ll let you know through the course of this review, but what I can tell you definitively is that the GeForce GTX 1080 clearly sits alone at the top of the GPU world.
NVIDIA's Ansel Technology
“In-game photography” is an interesting concept. Not too long ago, it was difficult to just capture the user's direct experience with a title. Print screen could only hold a single screenshot at a time, which allowed Steam and FRAPS to provide a better user experience. FRAPS also made video more accessible to the end-user, but it output huge files and, while it wasn't too expensive, it needed to be purchased online, which was a big issue ten-or-so years ago.
Seeing that their audience would enjoy video captures, NVIDIA introduced ShadowPlay a couple of years ago. The feature allowed users to, not only record video, but also capture the last few minutes. It did this with hardware acceleration, and it did this for free (for compatible GPUs). While I don't use ShadowPlay, preferring the control of OBS, it's a good example of how NVIDIA wants to support their users. They see these features as a value-add, which draw people to their hardware.