<

Intel

Intel teased its dedicated laptop GPU earlier in October, and it’s now rolling out the technology in earnest. The chip designer has announced that the first laptops using its Iris Xe Max graphics chip are available in some form, including the previously revealed Acer Swift 3X, the ASUS VivoBook Flip TP470 and Dell’s Inspiron 15 7000 2-in-1. You can buy the Dell hybrid at Best Buy now in the US, although you’ll have to wait for the Acer and ASUS machines to make their way stateside.

Iris Xe Max is, as you might guess, Intel’s answer to the low-end GPUs you sometimes find in thin-and-light portables. The company promises “great” 1080p gaming for popular titles like Hitman 2 and Metro Exodus (though it won’t compete against systems with beefier GPUs). It’s also smart enough to switch to integrated graphics when they might be more effective — Dota 2 runs slightly faster on the built-in Iris Xe than it does either the Xe Max or NVIDIA’s MX350, Intel claimed.

However, it might be more appealing beyond games. New Deep Link tech lets software split tasks between Intel’s integrated GPU on Tiger Lake-based PCs and the Iris Xe Max, sometimes delivering performance that would normally require a desktop video card. Intel is claiming twice the single-stream video encoding prowess of a GeForce RTX 2080 by early 2021, although we’ll have to see how well that translates in the real world. The Xe Max can devote all its power resources to the CPU for apps that depend more on conventional processing.

You’ll likely see more laptops, but Intel has also teased partnerships to roll out budget desktops using the Intel Xe Max’s underlying architecture (Xe-LP) in the first half of 2021. You should truly gaming-ready desktop GPUs from Intel next year, as well. In that light, Iris Xe Max is really laying groundwork for a more ambitious entry into the graphics space. Whether or not everyday users are receptive is another matter. The GPU market has been notoriously reluctant to accommodate more than two entrants, and Intel is still synonymous with slow graphics for many users.

<

 Teraflops have been a popular way to measure "graphical power" for years. The term refers to the number of calculations a GPU can perform, but while it’s been on spec sheets forever, more recently the teraflop has gone mainstream, appearing in marketing messages found in the launch of consoles like the Xbox Series X. With GPU core counts reaching five figures, it’s nice to have a simple point of comparison. Unfortunately, teraflops have never been less useful.

The term teraflop comes from FLOPs, or "floating-point operations per second," which simply means “calculations that involve decimal points per seconds.” Tera means trillion, so put together teraflops means “trillion floating-point operations per second.”

The most popular GPU among Steam users today, NVIDIA's venerable GTX 1060, is capable of performing 4.4 teraflops, the soon-to-be-usurped 2080 Ti can handle around 13.5 and the upcoming Xbox Series X can manage 12. These numbers are calculated by taking the number of shader cores in a chip, multiplying that by the peak clock speed of the card and then multiplying that by the number of instructions per clock. In contrast to many figures we see in the PC space, it's a fair and transparent calculation, but that doesn’t make it a good measure of gaming performance.

Almost every GPU family arrives with these generational gains

AMD’s RX 580, a 6.17-teraflop GPU from 2017, for example, performs similarly to the RX 5500, a budget 5.2-teraflop card the company launched last year. This sort of "hidden" improvement can be attributed to many factors, from architectural changes to game developers making use of new features, but almost every GPU family arrives with these generational gains. That's why the Xbox Series X, for example, is expected to outperform the Xbox One X by more than the “12 versus 6 teraflop” figures suggest. (Ditto for the PS5 and the PS4 Pro.)

The point is that, even within the same GPU company, with each year, changes in the ways chips and games are designed make it harder to discern what exactly "a teraflop" means to gaming performance. Take an AMD card and an NVIDIA card of any generation and the comparison has even less value.

All of which brings us to the RTX 3000 series. These arrived with some truly shocking specs. The RTX 3070, a $500 card, is listed as having 5,888 cuda (NVIDIA’s name for shader) cores capable of 20 teraflops. And the new $1,500 flagship card, the RTX 3090? 10,496 cores, for 36 teraflops. For context, the RTX 2080 Ti, as of right now the best "consumer" graphics card available, has 4,352 "cuda cores.” NVIDIA, then, has increased the number of cores in its flagship by over 140 percent, and its teraflops capability by over 160 percent.

Well, it has, and it hasn’t.

NVIDIA cards are made up of many "streaming multiprocessors," or SMs. Each of the 2080 Ti's 68 "Turing" SMs contain, among many other things, 64 "FP32" cuda cores dedicated to floating-point math and 64 "INT32" cores dedicated to integer math (calculations with whole numbers). 

The big innovation in the Turing SM, aside from the AI and ray-tracing acceleration, was the ability to execute integer and floating-point math simultaneously. This was a significant change from the prior generation, Pascal, where banks of cores would flip between integer and floating-point on an either-or basis.

NVIDIA

The RTX 3000 cards are built on an architecture NVIDIA calls "Ampere," and its SM, in some ways, takes both the Pascal and the Turing approach. Ampere keeps the 64 FP32 cores as before, but the 64 other cores are now designated as "FP32 and INT32.” So, half the Ampere cores are dedicated to floating-point, but the other half can perform either floating-point or integer math, just like in Pascal.

With this switch, NVIDIA is now counting each SM as containing 128 FP32 cores, rather than the 64 that Turing had. The 3070's "5,888 cuda cores" are perhaps better described as "2,944 cuda cores, and 2,944 cores that can be cuda."

As games have become more complex, developers have begun to lean more heavily on integers. An NVIDIA slide from the original 2018 RTX launch suggested that integer math, on average, made up about a quarter of in-game GPU operations.

The downside of the Turing SM is the potential for under-utilization. If, for example, a workload is 25-percent integer math, around a quarter of the GPU’s cores could be sitting around with nothing to do. That’s the thinking behind this new semi-unified core structure, and, on paper, it makes a lot of sense: You can still run integer and floating-point operations simultaneously, but when those integer cores are dormant, they can run floating-point instead.

[This episode of Upscaled was produced before NVIDIA explained the SM changes.]

At NVIDIA's RTX 3000 launch, CEO Jensen Huang said the RTX 3070 was "more powerful than the RTX 2080 Ti." Using what we now know about Ampere's design, integer, floating-point, clock speeds and teraflops, we can see how things might pan out. In that “25-percent integer” workload, 4,416 of those cores could be running FP32 math, with 1,472 handling the necessary INT32. 

Coupled with all the other changes Ampere brings, the 3070 could outperform the 2080 Ti by perhaps 10 percent, assuming the game doesn't mind having 8GB instead of 11GB memory to work with. In the absolute (and highly unlikely) worst-case scenario, where a workload is extremely integer-dependent, it could behave more like the 2080. On the other hand, if a game requires very little integer math, the boost over the 2080 Ti could be enormous.

Guesswork aside, we do have one point of comparison so far: a Digital Foundry video comparing the RTX 3080 to the RTX 2080. DF saw a 70 to 90 percent lift across generations in several games that NVIDIA presented for testing, with the performance gap higher in titles that utilize RTX features like ray tracing. That range gives a glimpse of the sort of variable performance gain we’d expect given the new shared cores. It’ll be interesting to see how a larger suite of games behaves, as NVIDIA is likely to have put its best foot forward with the sanctioned game selection. What you won’t see is the nearly-3x improvement that the jump from the 2080’s teraflop figure to the 3080’s teraflop figure would imply.

With the first RTX 3000 cards arriving in weeks, you can expect reviews to give you a firm idea of Ampere performance soon. Though even now it feels safe to say that Ampere represents a monumental leap forward for PC gaming. The $499 3070 is likely to be trading blows with the current flagship, and the $699 3080 should offer more-than enough performance for those who might previously have opted for the “Ti.” However these cards line up, though, it’s clear that their worth can no longer be represented by a singular figure like teraflops.

< < NVIDIAOn Tuesday we’re expecting to see details about NVIDIA’s next generation of graphics cards, but some of the information may have leaked early. Tom’s Hardware points out a couple of tweets that claim to share renders for Zotac Gaming cards in the new 3000 series. One from @momomo_us shows eight models in the new line, with RTX 3070, 3080 and 3090 designations.That seems to back up other notes suggesting that there won’t be any Ti/Super variants available at launch, and an unsourced report from Videocardz that cited the same lineup and provided detailed specs. Their report claimed the high-end 3090 units will include 24GB of GDDR6X memory and 5,248 cores, with a total graphics power requirement of 350W.Another tweet showed off supposed renders of a new 3090 Zotac card, complete with a triple fan cooling setup and and three PCI slot design. A previous leak showed off its enormous size compared to a current RTX 2080, while NVIDIA engineers confirmed a new 12-pin PCIe connector design in their video earlier this week. Whatever the specs are, we should find out more about them — and all the companies offering cards based on the new Ampere tech — on September 1st to celebrate 21 years of GeForce tech.

<

 

NVIDIA

On Tuesday we’re expecting to see details about NVIDIA’s next generation of graphics cards, but some of the information may have leaked early. Tom’s Hardware points out a couple of tweets that claim to share renders for Zotac Gaming cards in the new 3000 series. One from @momomo_us shows eight models in the new line, with RTX 3070, 3080 and 3090 designations.

That seems to back up other notes suggesting that there won’t be any Ti/Super variants available at launch, and an unsourced report from Videocardz that cited the same lineup and provided detailed specs. Their report claimed the high-end 3090 units will include 24GB of GDDR6X memory and 5,248 cores, with a total graphics power requirement of 350W.

Another tweet showed off supposed renders of a new 3090 Zotac card, complete with a triple fan cooling setup and and three PCI slot design. A previous leak showed off its enormous size compared to a current RTX 2080, while NVIDIA engineers confirmed a new 12-pin PCIe connector design in their video earlier this week. Whatever the specs are, we should find out more about them — and all the companies offering cards based on the new Ampere tech — on September 1st to celebrate 21 years of GeForce tech.

< <Right here’s a brand new set of slides being handed round that supposedly showcase AMD’s upcoming Radeon 6900 XT. They’re entirely false. Right here’s how you can tell:in the first slide, the branding has been up to date on the upper correct, but the branding on the genuine GPU hasn’t been. Also, that’s a Radeon 5700 cooler with a Vega water-cooler subsequent to it, and there’s a transparent flaw within the image the place the radiator attaches to the cardboard.The specs themselves are pretty affordable. I’m now not pronouncing how accurate I think they're, however the specs are the only a part of the slide that isn’t immediately fake. On the very least, I’d need to get out a calculator and run some numbers first.The fact that there’s a price on the cardboard is a further manner you understand this slide is false. Fee is normally the last factor a company decides on.This slide made me laugh out loud after I noticed it. Whoever created this work of art has by no means, ever, talked to anyone in advertising.Advertising, my friends, is all about optimism. You would word, for example, that after AMD declared Ryzen would have an IPC 1.4x bigger than Excavator, they didn't achieve this with a large slide labeled NO more BULLDOZENTThere’s no method in hell AMD would ever advertise RDNA2 in a fashion that implied RDNA or any prior GPU used to be “compromised.” AMD continues to be shipping Vega silicon in its APUs and as a part of its compute trade.AMD is unlikely to call its ray tracing implementation “RXRT,” and the estimated efficiency influence of enabling the function is hilarious, to put it flippantly.I was no longer in particular delighted with Turing when it came out and i count on each Ampere and RDNA2 to offer sophisticated efficiency in ray tracing workloads, but there’s no danger that AMD takes a 5-9 percent penalty for enabling ray-tracing results. If real-time ray tracing most effective carried a 5-10 percent performance penalty relative to rasterization, we’d have integrated RTRT a very long time ago.The average performance hit for utilising ray tracing is more alongside the strains of fifty-80 percentage. When you consider that we best have one new release of hardware from one company, we don’t know the way a lot that number can also be extended — nevertheless it beggars perception to consider AMD has reduce it by practically an order of magnitude. It’s no longer even clear which RTX video games will aid RDNA2 ray tracing out of the gate.The “best 4K Gaming experience” is spot-on for a advertising slide, however AMD doesn’t give you recreation element settings in the bars of its slides, and they don’t label up the y-axis to the factor that you must twist your neck like an owl with the intention to learn it.Also, excellent to peer top-notch efficiency in red lifeless Redemption, a game that on no account obtained a computer unlock. By no means heard of “Witcher three,” either, when you consider that the precise identify of the series is “The Witcher.”that is what it looks like when men and women with more aspiration than Photoshop attempt to troll AMD lovers. There’s a constant design language to AMD’s slide decks and a regular manner that businesses be in contact about their merchandise. These slides fail at both.I declare this GPU a brand new and distinct product under the solar. Imparting the AMD Radeaint 6900 XT, launching September 2020.Function photograph is the AMD Radeon 5700 XT.

<

Right here’s a brand new set of slides being handed round that supposedly showcase AMD’s upcoming Radeon 6900 XT. They’re entirely false. Right here’s how you can tell:

in the first slide, the branding has been up to date on the upper correct, but the branding on the genuine GPU hasn’t been. Also, that’s a Radeon 5700 cooler with a Vega water-cooler subsequent to it, and there’s a transparent flaw within the image the place the radiator attaches to the cardboard.

The specs themselves are pretty affordable. I’m now not pronouncing how accurate I think they're, however the specs are the only a part of the slide that isn’t immediately fake. On the very least, I’d need to get out a calculator and run some numbers first.

The fact that there’s a price on the cardboard is a further manner you understand this slide is false. Fee is normally the last factor a company decides on.

This slide made me laugh out loud after I noticed it. Whoever created this work of art has by no means, ever, talked to anyone in advertising.

Advertising, my friends, is all about optimism. You would word, for example, that after AMD declared Ryzen would have an IPC 1.4x bigger than Excavator, they didn't achieve this with a large slide labeled NO more BULLDOZENT

There’s no method in hell AMD would ever advertise RDNA2 in a fashion that implied RDNA or any prior GPU used to be “compromised.” AMD continues to be shipping Vega silicon in its APUs and as a part of its compute trade.

AMD is unlikely to call its ray tracing implementation “RXRT,” and the estimated efficiency influence of enabling the function is hilarious, to put it flippantly.

I was no longer in particular delighted with Turing when it came out and i count on each Ampere and RDNA2 to offer sophisticated efficiency in ray tracing workloads, but there’s no danger  that AMD takes a 5-9 percent penalty for enabling ray-tracing results. If real-time ray tracing most effective carried a 5-10 percent performance penalty relative to rasterization, we’d have integrated RTRT a very long time ago.

The average performance hit for utilising ray tracing is more alongside the strains of fifty-80 percentage. When you consider that we best have one new release of hardware from one company, we don’t know the way a lot that number can also be extended — nevertheless it beggars perception to consider AMD has reduce it by practically an order of magnitude. It’s no longer even clear which RTX video games will aid RDNA2 ray tracing out of the gate.

The “best 4K Gaming experience” is spot-on for a advertising slide, however AMD doesn’t give you recreation element settings in the bars of its slides, and they don’t label up the y-axis to the factor that you must twist your neck like an owl with the intention to learn it.

Also, excellent to peer top-notch efficiency in red lifeless Redemption, a game that on no account obtained a computer unlock. By no means heard of “Witcher three,” either, when you consider that the precise identify of the series is “The Witcher.”

that is what it looks like when men and women with more aspiration than Photoshop attempt to troll AMD lovers. There’s a constant design language to AMD’s slide decks and a regular manner that businesses be in contact about their merchandise. These slides fail at both.

I declare this GPU a brand new and distinct product under the solar. Imparting the AMD Radeaint 6900 XT, launching September 2020.

Function photograph is the AMD Radeon 5700 XT.