Hands On With The AMD Radeon RX 6600 XT

[artivision]

Actually, that's not corrrect. The low 75W> TDP cards can actually be way more powerful than iGPU's. For example, a long time ago I had the 2012 Radeon HD 7750, which is around 60-70W TDP, and it could easily run many triple A games of that time at high settings, which is way beyond what iGPU's could achieve at least then. I think there is no reason Nvidia / AMD can't make a 75W> card today that is way better than an iGPU. It's simply because both companies completely neglected midrange / entry level markets. I mean after 4+ years there is still no worthy successor to the midrange RX 580 / GTX 1060.

RX 5500 xt undervolt overclock, covers you with 85w.

[Linuxxx]

1080p GPU with 3 FANs... When will I have a chance to replace my R9 380 ITX Compact?

Incidentally I also own an AMD R9 380 [4 GB], though non-ITX and therefore most likely a little bit more powerful than your version.

And depending on your preferences you may not like this notion, but I can tell you from first-hand experience that a nVidia 1650 [GDDR6] is actually faster than my R9 380.

I know this because of my custom mpv.conf, where I had to slightly lower the 'error-diffusion' algorithm used on my AMD card because it couldn't keep up where the 1650 wasn't struggling, even though it draws considerably less power while delivering the better performance of these two cards.

On both setups I used the OpenGL output, because RadeonSI would consistenly outperform the Vulkan option with the RADV driver the last time I tested this, which must be around half-a-year by now.

So as long as you are fine with running a binary blob, nVidia right now has the only viable option for a 75 Watts GPU, unfortunately.

Hopefully Intel really delivers with their dGPU line-up, since competition is direly needed here!

[birdie]

You all missed the point of my post, so I'll be more clear. Where is the Radeon RX6400 with 75W TBP? Or even better, an RX6200 with 35-50W TBP?

AMD reckons the built-in Vega graphics in its Ryzen APUs is good enough and that's my opinion as well. Now tons of people don't have a built-in GPU and that's a valid concern but such people normally know what they are doing and buy beefy GPUs instead.

[moilami]

Nice to see not only Linux support printed in the box, but it be in alphabetical order, not put dead last like a leper you pretend that didn't exist.

To say the least.

[leipero]

Agree that the HD 7750/7770 (Cape Verde IIRC) was faster than the iGPUs of the time, but I expect that the integrated GPUs in Renoir and Cezanne would be faster than the HD 7750 these days, even with shared and slightly slower memory.

There has been a lot of discussion about the lowest tier cards in other threads so I won't repeat it all but:

#1 - yes there is still a slot between APU and "great 1080p dGPU"

#2 - problem is that the market for those boards is much smaller these days and barely makes sense financially

The product would basically be "laptop dGPU stuck on a PCIE card" but that seems to be a tough sell to the board partners. I like the idea of a no-power-connector mini-ITX size card and own a few of them but that seems to be a minority view.

In fairness, the RX 5500XT was a bit faster, a bit cheaper and used less power. It just wasn't exciting I guess.

That's an even smaller market, unfortunately... while the cost of making new chips is going through the roof.

What kind of use cases do you see for graphics acceleration on the server ? My impression was that the GPUs integrated into BMCs were doing a decent job of covering user needs.

For newer titles even Vega 3 can be close to 7750/70 under specific conditions, and Vega 8 even superior to them, while in older, depending on the freqency it runs at, it can be about the same or better.

I think cache for iGPUs is an interesting option (as long as it's not expensive), because you don't need super fast memory like "gaming cache" (L3 on Zen3), assuming game have proper optimization for texture loading, such cache could make wonders, and because it doesn't need to be super fast, cheaper (and larger) can be used, even in systems with DDR5 and so on, there would be benefit of it IMO.
Such APUs would solve the issue of low end GPUs, ofc. with power limitations on how much one could do not to stress motherboard manufacturers, because games have different requirements in 2021 in comparison to 2011, drop the resolution and you solved most of the problems, add to that good upscaling solution, and you are done.

Usually at office, Intel CPUs are used (mostly i5 range or so), why? In reality because office computers do not need that much compute power as long as they operate well enough for the OS and applications (office, browsers etc.), so people don't upgrade them for years/decade. But another reason where AMD is potentially missing is because for every business owner, less investment = better, and all of those Intel CPUs have iGPUs, and none of those machines have (or need) dedicated GPU, so overall cost of say 20 PCs, even with more expensive boards and CPUs would be lower for that purpose on agregate in comparison to the equvalent X with similar compute power with dedicated GPU (cheapest one).
Another thing is, iGPUs can come up handy for troubleshooting, and can make a difference for average users between functional and non-functional PC when dGPU is under RMA or something like that. Something like "Vega 1-3" would be sufficient I guess, some basic iGPU that requires little space and is possible to produce and include in price without net-loss at the very least and without motherboard being overcomplicated to keep the price as it is. Arguments to consider.

[bridgman]

Usually at office, Intel CPUs are used (mostly i5 range or so), why? In reality because office computers do not need that much compute power as long as they operate well enough for the OS and applications (office, browsers etc.), so people don't upgrade them for years/decade. But another reason where AMD is potentially missing is because for every business owner, less investment = better, and all of those Intel CPUs have iGPUs, and none of those machines have (or need) dedicated GPU, so overall cost of say 20 PCs, even with more expensive boards and CPUs would be lower for that purpose on agregate in comparison to the equvalent X with similar compute power with dedicated GPU (cheapest one).

Another thing is, iGPUs can come up handy for troubleshooting, and can make a difference for average users between functional and non-functional PC when dGPU is under RMA or something like that. Something like "Vega 1-3" would be sufficient I guess, some basic iGPU that requires little space and is possible to produce and include in price without net-loss at the very least and without motherboard being overcomplicated to keep the price as it is. Arguments to consider.

Yep - our CPUs for those markets also include GPUs, although our GPUs tend to be a bit more powerful than the corresponding Intel ones.

This isn't always obvious because we did not sell Renoir into the DIY desktop market - qualification takes a long time and Cezanne was running right behind Renoir - but we did sell a lot of 4xxxG parts to OEMs which went into the same office systems you describe.

We are shipping 5xxxG parts into the DIY market now, although the "most office-y version" (5300G) is currently OEM only.

What we don't have today is "tiny little GPUs for troubleshooting a gaming/workstation system", whether they be integrated or discrete, and opinions vary re: how important those are. Most DIY'ers tend to have an old dGPU sitting around but not all of them do. The sad thing is that there are probably tens of thousands of those cards scrapped every year but still functional.

On the other hand my old Vesa bus dGPUs are probably running off the end of their usefulness

[user1]

Have to disagree there - the RX 5500XT was a bit faster, a bit cheaper and used less power than either of those. Unfortunately even the 5500XT is selling for 4x MSRP or higher right now.

Well, to me, a worthy successor means that the card is much faster than the one it replaces and it's at least about the same price. I would say that the RX 5600xt was pretty close being a 580 / 1060 replacement (AFAIR, its MSRP was below 300$ and it was as fast as a GTX 1080) but yeah.. with todays prices because of obvious reasons, you just can't expect that.

[bridgman]

Well, to me, a worthy successor means that the card is much faster than the one it replaces and it's at least about the same price. I would say that the RX 5600xt was pretty close being a 580 / 1060 replacement (AFAIR, its MSRP was below 300$ and it was as fast as a GTX 1080) but yeah.. with todays prices because of obvious reasons, you just can't expect that.

I hate to say it, but I think those days are mostly gone. Those improvements in price/performance were largely made possible by ongoing reductions in cost per transistor from newer fab processes, but between 14/12nm and 8/7nm those ongoing reductions largely went away (not completely but mostly).

The rule of thumb used to be that GPU cost was roughly proportional to die size and that moving to new fab processes gave the GPU vendors more transistors to work with (and hence more performance) at the same price point. These days the increasing complexity and cost of new fab processes largely matches the increased transistor density, and as a consequence GPU cost tracks number of transistors more closely than it tracks die size.

Remember all those years when people were saying that the semiconductor industry would hit a wall and be unable to shrink fab processes past a certain point ? It's probably fair to say that rather than turning out to be impossible it turned out to be extremely complicated and expensive. My low-quality understanding is that the move from planar FETs to Fin-FETs was a big contributor to the complexity.

TL;DR - unless/until something changes in the fab world the degree of price/performance improvement with each new fab process dropped substantially in the last 5 years (even before shortages and COVID-19) and shows no sign of coming back.

Improvements in performance and power/performance are still happening, but price/performance not so much.

[bridgman]

1080p GPU with 3 FANs... When will I have a chance to replace my R9 380 ITX Compact?

https://www.asrock.com/Graphics-Card/AMD/Radeon%20RX%206600%20XT%20Challenger%20ITX%208GB/

I believe this is the only single fan ITX 6600XT at the moment; most cards seem to have 2 or 3 fans.

[leipero]

Yep - our CPUs for those markets also include GPUs, although our GPUs tend to be a bit more powerful than the corresponding Intel ones.

This isn't always obvious because we did not sell Renoir into the DIY desktop market - qualification takes a long time and Cezanne was running right behind Renoir - but we did sell a lot of 4xxxG parts to OEMs which went into the same office systems you describe.

We are shipping 5xxxG parts into the DIY market now, although the "most office-y version" (5300G) is currently OEM only.

What we don't have today is "tiny little GPUs for troubleshooting a gaming/workstation system", whether they be integrated or discrete, and opinions vary re: how important those are. Most DIY'ers tend to have an old dGPU sitting around but not all of them do. The sad thing is that there are probably tens of thousands of those cards scrapped every year but still functional.

On the other hand my old Vesa bus dGPUs are probably running off the end of their usefulness

I see your point, still, those GPUs in those APUs are way overkill for office work (for now at least). Depending on the owner (and place), sometimes it's actually cheaper to purpose build PCs than use OEM ones, so I was refering more to that. Also, what such business owners prefer is how easy is to set and maintain the system, so having "easy interface all in one" driver (Radeon Software is unnecessarily heavy for such use on what is usually Windows env., I don't know if it's possible to install driver only from GUI, I assume it is tho.) for such systems would be beneficial IMO. I don't know if what I'm saying is possible from the cost/benefit perspective, but if it is, I see no reason why not to do it.

That's sad thing, but, it's how people work in general, for those office PCs those cards could be useful even today indeed .