This test is another in a series of studies to learn more about nVidia’s new Volta architecture. Although Volta in its present form is not the next-generation gaming architecture, we would anticipate that key performance metrics can be stripped from Volta and used to extrapolate future behavior of nVidia’s inevitable gaming arch, even if named differently. One example would be our gaming benchmarks, where we observed significant performance uplift in games leveraging asynchronous compute pipelines and low-level APIs. Our prediction is that nVidia is moving toward a future of heavily support asynchronous compute job queuing, where the company is presently disadvantaged versus its competition; that’s not to say that nVidia doesn’t do asynchronous job queuing on Pascal (it does), but that AMD has, until now, put greater emphasis on that particular aspect of development.
This, we think, may also precipitate more developer movement toward these more advanced programming techniques. With the only two GPU vendors in the market supporting lower level APIs and asynchronous compute with greater emphasis, it would be reasonable to assume that development would follow, as would marketing development dollars.
In this testing, we’re running benchmarks on the nVidia Titan V to determine whether GPU core or memory (HBM2) overclocks have greater impact on performance. For this test, we’re only using a few key games, as selected from our gaming benchmarks:
- Sniper Elite 4: DirectX 12, asynchronous compute-enabled, and showed significant performance uplift in Volta over Pascal. Sniper responds to GPU clock changes in drastic ways, we find. This represents our async titles.
- Ashes of the Singularity: DirectX 12, but less responsive than Sniper. We were seeing ~10% uplift over the Titan Xp, whereas Sniper showed ~30-40% uplift. This gives us a middle-ground.
- Destiny 2: DirectX 11, not very responsive to the Titan V in general. We saw ~4% uplift over the Titan Xp at some settings, though other settings combinations did produce greater change. This gives us a look at games that don’t necessarily care for Volta’s new async capabilities.
We are also using Firestrike Ultra and Superposition, the latter of which is also fairly responsive to the Titan’s dynamic ray-casting performance.
We are running the fan at 100% for all tests, with the power offset at 120% (max) for all tests. Clocks are changed according to their numbers in the charts.
As we work toward our inevitable hybrid mod on the nVidia Titan V, we must visit the usual spread of in-depth thermal, power, and clock behavior testing. The card uses a slightly modified Titan Xp cooler, with primary modifications found in the vapor chamber’s switch to copper heatfins. That’s the primary change, and not one that’s necessarily all that meaningful. Still, the card needs whatever it can get, and short of a complete cooler rework, this is about the most that can fit on the current design.
In this Titan V benchmark, we’ll be looking at the card’s power consumption during various heavy workloads, thermal behavior of the MOSFETs and GPU core, and how frequency scales with thermals and power. The frequency scaling is the most important: We’ve previously found that high-end nVidia cards leave noteworthy performance (>100MHz boost) on the table with their stock coolers, and suspect the same to remain true on this high-wattage GPU.
The nVidia Titan V is not a gaming card, but gives us some insights as to how the Volta architecture could react to different games and engines. The point here isn’t to look at raw performance in a hundred different titles, but to think about what the performance teaches us for future cards. This will teach us about the Volta architecture; obviously, you shouldn’t be spending $3000 to use a scientific card on gaming, but that doesn’t mean we can’t learn from it. Our tear-down is already online, but now we’re focusing on Titan V overclocking and FPS benchmarks, and then we’ll move on to production, power, and thermal content.
This nVidia Titan V gaming benchmark tests the Volta architecture versus Pascal architecture across DirectX 11, DirectX 12, Vulkan, and synthetic applications. We purchased the Titan V for editorial purposes, and will be dedicating the next few days to dissecting every aspect of the card, much like we did for Vega: Frontier Edition in the summer.
After completing all of our gaming, power, thermal, and other benchmarks for the new nVidia Titan V graphics card, we took the unit apart for cooler, PCB, and VRM analysis. We’ll be joined by overclocker ‘Buildzoid’ in the next few days for the advanced overclocking analysis of the PCB and VRM, but have some immediate information on the assembly of the Titan V and its cooler.
The card follows the same screw pattern as all previous nVidia Founders Edition cards, including the Titan Xp and GTX 1080, primarily isolating its cooler and shroud into a single, separable unit. Build materials are all the same, assembly is the same, but the underlying GPU, HBM2, VRM, and heatsink are different.
Vega 64 may consume more power than a GTX 1080, but until now, we haven’t known if that impact is relevant to room temperature. That’s what we wanted to know, and we eventually expanded that concept to include how much a 900W+ mining machine increases room temperature, a 600W machine, and so on. We were able to effectively replace any need of a heater for the past week, and right when it started to get colder.
In this test, we’re looking at the room ambient impact of various PC builds. This helps to conceptualize the real-world impact of all those power and thermal tests you see us (and others) publish, as it puts real numbers to the user experience outside of the case. Although this concept has about a million variables and “what ifs,” we controlled to the best of our abilities, are laying-out all the major variables, and can present an academic experiment that demonstrates room temperature increase from computer equipment. All watts are basically created equal, for the purposes of this test: A 940W mining rig will output just as much heat into the room as a 940W gaming rig, or a 940W rendering machine, and so forth; as long as the power load is equal between all of these (read: constant), watts are watts, and you can extrapolate room temperature for each type of machine.
The testing originally was concepted after our Vega 56 Hybrid mod, which used power mods and other mods to push the card up towards 400W of power consumption. We wanted to test a straight Vega 56 versus GTX 1070 for room ambient impact, but shifted that up a tier (to Vega 64 and a GTX 1080) for some parts that are more likely to show a difference. After that, we shifted up to a 940W mining machine, then picked a middle-ground ~600W machine (which could also represent SLI gaming or HEDT render systems).
This episode of Ask GN headlines with answering the most common question we’ve seen in the past 24 hours: Should I buy now or wait for Volta? That’ll start us off for this episode, followed by clarification of VRM quality, a history lesson on AM4 motherboards at launch and HIS existence, and silicon death from overclocking. This episode runs about 25 minutes, with each question timestamped within the video. We also have the timestamps and questions marked below, if you’d like to see when a particular topic of interest appears.
The Volta topic, we think, is among the most interesting and common for questions right now. This seems to come around for every new architecture, and our answers are generally the same. Find out more below!
NVidia introduced its new Titan V GPU, which the company heralds as the “world’s most powerful GPU for the PC.” The Titan V graphics card is targeted at scientific calculations and simulation, and very clearly drops any and all “GTX” or “gaming” branding.
The Titan V hosts 21.1B transistors (perspective: the 1080 Ti has 12B, P100 has 15.3B), is capable of driving 110TFLOPS of Tensor compute, and uses the Volta GPU architecture. We are uncertain of the lower level specs, and do not presently have a block diagram for the card. We have asked for both sets of data.
AMD’s partner cards have been on hold for review for a while now. We first covered the Vega 64 Strix when we received it, which was around October 8th. The PowerColor card came in before Thanksgiving in the US, and immediately exhibited similar clock reporting and frequency bugginess with older driver revisions. AMD released driver version 17.11.4, though, which solved some of those problems – theoretically, anyway. There are still known issues with clock behavior in 17.11.4, but we wanted to test whether or not the drivers would play nice with the partner cards. For right now, our policy is this: (1) We will review the cards immediately upon consumer availability or pre-order, as that is when people will need to know if they’re any good; (2) we will review the cards when either the manufacturer declares them ready, or at a time when the cards appear to be functioning properly.
This benchmark is looking at the second option: We’re testing whether the ASUS Strix Vega 64 and PowerColor Red Devil 64 are ready for benchmarking, and looking at how they match versus the reference RX Vega 64. Theoretically, the cards should have slightly higher clocks, and therefore should perform better. Now, PowerColor has set clock targets at 1632MHz across the board, but “slightly higher clocks” doesn’t just mean clock target – it also means power budget, which board partners have control over. Either one of these, particularly in combination with superior cooling, should result in higher sustained boost clocks, which would result in higher framerates or scores.
We need some clarity on this issue, it seems.
TLDR: Some AMD RX 560 graphics cards are selling with 2 CUs disabled, resulting in 896 streaming processors to the initially advertised 1024 (64 SPs per CU). Here’s the deal: That card already exists, and it’s called an RX 460; in fact, the first two lines of our initial RX 560 review explicitly states that the driving differentiator between the 460 and 560, aside from the boosted clocks, was a pre-enabled set of 2CUs. The AMD RX 460s could already be unlocked to have 16 CUs, and the RX 560 was a card that offered that stock, rather than forcing a VBIOS flash and driver signature.
The RX 560 with 2CUs disabled, then, is not a new graphics card. It is an RX 460. We keep getting requests to test the “new” RX 560 versus the “old” RX 560 with 1024 SPs. We already did: The RX 560 review contains numbers versus the RX 460, which is (literally) an RX 560 14CU card. It is a rebrand, and that’s likely an attempt to dump stock for EOY.
Jon Peddie Research reports that the AIB market is likely returning to normal seasonal trends, meaning the market will be flat or moderately down from Q4 2017 through Q1 2018.
In a typical year, the AIB market is flat/down in Q1, down in Q2, up in Q3, and flat/up in Q4. The most dramatic change is usually from Q2 to Q3, on average a 14.4% increase (over the past 10 years). Q3 2016 was roughly twice that average with more than 15 million AIBs shipped, 29.1% more than Q2 and a 21.5% increase year-over-year.
We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.