News AMD deep-dives Zen 5 — Ryzen 9000 and AI 300 benchmarks, Zen 5, RDNA 3.5 GPU, and XDNA 2 microarchitectures

[usertests]

The 105W 7800X3D uses more like 50W when gaming, so the 9700X slide is questionable.

Ryzen 7 7800X3D: Ahead of Core i9 for gaming at 40% power draw - Page 36 of 41 - HWCooling.net

Average processor power drawThe most popular Ryzen 7000 with 3D V-cache obliterates competition in its targeted environment. Intel doesn’t measure up to the Ryzen 7 7800X3D in a comparable price range in terms of speed, and even if someone disregards price and sees the Core i9-13900K(S) as a an...

www.hwcooling.net

Strix Point performance vs. desktop cards... RX 6400? GTX 970? Seems it's around that, and the memory optimizations help it even if it's not getting significantly faster memory.

 

[Guardians Bane]

Intel's never bad, wrong, always better. And when somebody proves you wrong or misinterpreting then you simply exercise goalpost shifting. Excuse me for ad personam commentary, but it's kinda funny that you tend to bend each story to "Intel is better

I've noticed that as well .. almost as bad as the comment section on WCCF Tech and the Intel fans.

 

[usertests]

I've noticed that as well .. almost as bad as the comment section on WCCF Tech and the Intel fans.

You haven't been there often enough. If this comment section is a playground fight, WccfKek is the Vietnam War.

 

[bit_user]

Also testing was done with a ~ $200 water cooler....just saying but if you need that to run the 9950x at stock, 230w ppt, then it's gonna be a joke, at least on the intel system you would get like ~350W out of that.

"Dear concerned tech enthusiast. AMD has heard your sincere and heartfelt opinions about the cooling efficiency of our recent products and hopes our improved thermal solution meets with your approval."

\s

​

Don't tell me you didn't even look through the slides! How can you criticize what you don't understand?

 

[bit_user]

BTW, the font kerning on some of these slides is just atrocious. Some letters are overlapping, while others are spaced so far apart that it almost looks like there's a space where there's not!

Here's one such example (but I'm sure not even the worst):

​

At the top, the letters in "Gen" are all overlapping, while "Strix" looks more like "St rix". At the bottom, "heterogeneous" looks more like "het erogeneous".

I'm not normally one to complain about such things, but it's so bad that it really affects their readability. So, um, like what gives, AMD??

 

[Neilbob]

BTW, the font kerning on some of these slides is just atrocious. Some letters are overlapping, while others are spaced so far apart that it almost looks like there's a space where there's not!

I'm glad it's not just me who noticed that, but didn't want to appear pernickety by pointing it out (don't know why - I really don't care about being pernickety); luckily, you beat me to it. This almost appears as if a different font was used, and then accidentally changed to what looks like some variation Arial, except it happens in every single slide so I'm assuming this was the intended typeface.

I used to be a sort-of graphic designer/desktop publishing guy, and I was almost obsessed with the kerning/spacing of text in headings and sub-headings. Not doing so just appears unprofessional. That's what these slides look like.

 

[-Fran-]

maybe waiting how bad intel is hurting with 14900k and arrow lakes pricing?

Maybe not an outcome of the issues with Raptor Lake, but trying to read the tea leaves in regards to Arrow Lake and Meteor Lake (perf-wise), as they'll be the real competition for Ry9K. I'm sure they want to squeeze as much money as possible before they launch while they can make use of Intel's current bad press around consumer. Which is annoying, but corp do corp things, I guess.

All in all, I think you're right, but I wanted to give more of my "feelings" into it.

Regards.

 

[mitch074]

I'm glad it's not just me who noticed that, but didn't want to appear pernickety by pointing it out (don't know why - I really don't care about being pernickety); luckily, you beat me to it. This almost appears as if a different font was used, and then accidentally changed to what looks like some variation Arial, except it happens in every single slide so I'm assuming this was the intended typeface.

I used to be a sort-of graphic designer/desktop publishing guy, and I was almost obsessed with the kerning/spacing of text in headings and sub-headings. Not doing so just appears unprofessional. That's what these slides look like.

PDF with bad replacement font, or bad format conversion - if each glyph's positioning is saved instead of that of a text box, and then the font - not included in the PDF - is replaced with one that has strongly different metrics, that's what you'd get.

 

[stuff and nonesense]

Funny thing is that I haven't heard anyone talking about AMD Ryzen CPUs failing due to exceeding power / voltage / current limits. And yet, the Core i9-13900K and 14900K are apparently becoming a serious issue a year or two after release. I had to get my 13900K chip replaced because it failed.

Anyway, forum regulars know Terry has a massively skewed pro-Intel / anti-AMD perspective. It's not worth trying to argue facts with such types in my experience.

I’m waiting to see more from Level 1 Techs. In Wendell’s research he has uncovered i9s failing in a server environment without overclocking and with memory dialled back to sub 5000mt for stability.

If this proves to be valid then there is far more going on than simple overheating.

 

[KyaraM]

T

Funny thing is that I haven't heard anyone talking about AMD Ryzen CPUs failing due to exceeding power / voltage / current limits. And yet, the Core i9-13900K and 14900K are apparently becoming a serious issue a year or two after release. I had to get my 13900K chip replaced because it failed.

Anyway, forum regulars know Terry has a massively skewed pro-Intel / anti-AMD perspective. It's not worth trying to argue facts with such types in my experience.

That seems to not be the full issue, though, seeing how even chips with stock limits or reduced clocks set to 5.5GHz/undervolted etc. seem to degrade, only slower, as mentioned elsewhere. It probably contributes, yes, but at least to me it looks like it's not the full story. GN and L1 Tech have great videos on the topic, looking at CPUs used in entry-level servers. And stock Intel limits aren't that far out there with the affected chips compared to both AMD and ALD, and those don't have issues (so far...). Actually, power limits for the 13900K are the exact same as for the 12900K, which doesn't have these issues.

Mind, I'm not saying Terry is right, or that you are completely wrong, just that I feel it might be a bit simplistic to blame power limits etc when we don't know what really is going on...

 

[KyaraM]

I’m waiting to see more from Level 1 Techs. In Wendell’s research he has uncovered i9s failing in a server environment without overclocking and with memory dialled back to sub 5000mt for stability.

If this proves to be valid then there is far more going on than simple overheating.

He makes it explicitly clear those server CPUs are not overheating, so that is absolutely not the issue here, yeah. Unless they heat up catastrophically for micro seconds that cannot be detected...

 

[-Fran-]

I was reading more into their "IHS" improvement and it's just a bunch of smokes and mirrors it seems. The biggest improvement seems to not come from a better "IHS" itself, but rather "clever" thermal sensor placement, which is just another way of saying "we're hiding the temps better this time"... Ugh...

Well, I'll try to give them the benefit of the doubt and hope the temps reported by those sensors are actually correct and with Ry7K the temps have been higher than actual values (because that's the obvious consequence of what they've said).

I hope peeps doing the testing put a close eye on this, since it's not a minor thing to me 😀

Regards.

 

[KyaraM]

I was reading more into their "IHS" improvement and it's just a bunch of smokes and mirrors it seems. The biggest improvement seems to not come from a better "IHS" itself, but rather "clever" thermal sensor placement, which is just another way of saying "we're hiding the temps better this time"... Ugh...

Well, I'll try to give them the benefit of the doubt and hope the temps reported by those sensors are actually correct and with Ry7K the temps have been higher than actual values (because that's the obvious consequence of what they've said).

I hope peeps doing the testing put a close eye on this, since it's not a minor thing to me 😀

Regards.

That would be kinda scummy if true tbh. Let's hope it's not the case...

 

[bit_user]

This almost appears as if a different font was used, and then accidentally changed to what looks like some variation Arial, except it happens in every single slide so I'm assuming this was the intended typeface.

After I posted that, I seemed to recall a minor controversy that arose after it was noticed that AMD was using a Mac laptop to play their slides at one of these types of events, a couple years ago. I wonder if the PR firm switched to a different software package they could run on Windows, that they weren't as comfortable with. Probably not, but just maybe one has to do with the other.

I was almost obsessed with the kerning/spacing of text in headings and sub-headings. Not doing so just appears unprofessional. That's what these slides look like.

Over the past couple decades of Linux use, I've seen varying issues like this in a few software packages. Not sure if it was from poor support for TrueType fonts, due to patent issues, or what. I'm also reminded of this:

https://www.phoronix.com/news/HarfBuzz-Hinting-Woe
​

 

[bit_user]

I was reading more into their "IHS" improvement and it's just a bunch of smokes and mirrors it seems. The biggest improvement seems to not come from a better "IHS" itself, but rather "clever" thermal sensor placement,

Could it possibly be a "both-and" situation, where they both moved the thermal sensor and made some TIM or IHS-related improvements, but the person claiming that was only aware of one and not the other?

I think it's not in their interest to have worse information about die temperatures. It seems likely to blow up in their faces, if you'll excuse the turn of phrase.

 

[TheHerald]

Comparing power draw to CPU die temp from different companies is worthless. The 14900k is VASTY more difficult to cool than the Ryzens due to its extra power draw.

No, intel cpus are vastly easier to cool. Vastly. They just have better heat dissipation. How do you think reviewers are able to hit 300 and 350w power draw out of these chips? It's because they are easy to cool

 

[TheHerald]

Funny thing is that I haven't heard anyone talking about AMD Ryzen CPUs failing due to exceeding power / voltage / current limits.

Uhm, have you heard about high vsoc making amd cpus immolate and cook the mobo alongside them? It wasn't too long ago.

 

[stuff and nonesense]

He makes it explicitly clear those server CPUs are not overheating, so that is absolutely not the issue here, yeah. Unless they heat up catastrophically for micro seconds that cannot be detected...

Exactly, if they are not overheating then what is the underlying problem?

 

[bit_user]

Exactly, if they are not overheating then what is the underlying problem?

The internet is full of speculation and people who think they know. The latest word from Intel is they don't know, but are still looking into it!

https://www.tomshardware.com/pc-com...re-i9-crashing-issues-investigation-continues
​

Modern photolithography is complex business, pushing the boundaries of physics, chemistry, and manufacturing. I don't trust anyone who's not a lithography expert who thinks they know what's wrong. However, you can find some worthwhile advice on howto mitigate the risk or frequency of the issue.

 

[SonoraTechnical]

I was hoping maybe we see a refresh of the current Radeon discreete GPUs.... Perhaps a Radeon RX 7950 XTX w/ RDNA 3.5 .. since the 7950 was never released... perhaps it could be with the RDNA 3.5 enhancements?

Appears RDNA 4 (RX 8000 series cards) won't release anything that is faster faster than the current RX 7900 XTS (RNDA 3). A faster RX 7950 XTX w/ RDNA 3.5 tech would be a good holdover, no?

 

[Silas Sanchez]

It's not like it's a secret or in any way controversial that ryzen is very hard to cool.
The same amount of cooling that is required to get the PPT of ryzen at thermal throttle temps is enough to give intel 50% more power draw at 8 degrees lower temp.
https://www.anandtech.com/show/17641/lighter-touch-cpu-power-scaling-13900k-7950x/3

Putting aside that you clearly have some deep vested psychological interest in sprouting Sophistry.

High temps is the big unknown, of course the only way we will ever know is to simple wait and see what pans out. The experts dont know, they cant even tie their own shoe laces, i have seen lithium battery scientests involved with decades of studying batteries and they are clueless on all the tough realworld practical questions, such questions only the users have a chance of knowing.
Its up to the individual to do what they believe is best based on their current knowledge and future needs and take the necessary steps to protect oneself and the continuity of their rig. Never buy new hardware until its been out in the wild for a good half a year or else you risk being a victim of good ol fashion human incompetence.
I tend to measure success by how long the system lasts, so thats why I use powerful workstation laptops for basic work as they are gurenteed to be more than enough even when the thermal paste goes bad. I cant afford to get a new system in 5years as windows is officialy dead due to its excessive toxic rubbish inside it. Im stuck riding out win10.

For gamers who get ~5years out of a system id say worrying about heat isn't an issue, but if you want to get 10years plus of solid use then absolutely, don't ever listen to these companies, AMD said something like 'Laptops have been sitting at 90C+ for years'... a pretty fallacious load of crap. So I take the necessary steps to protect myself from corporate goons and only listen to the science which last I check was about as scant as tap water vs distilled water. So im making my own waves as i go, learning as i go, hoarding knowledge etc. I don't game, and that makes life alot easier and if I did id just go and get a Playstation for a fraction of the cost of a gpu, 4 grand for a gpu? um no...haha..of course, no... those criminals want me to pay that for gaming can put a real 8year warranty on it.

 

[abufrejoval]

Ugh... I don't like the fact that AMD has been hush hush with the prices. They've been forthcoming with them for a while, but now they are not. I don't want to read too much in between the lines, but I don't like it.

Prices are bound to be highly dynamic and hopefully not the direction they headed when the 5950X launched.

I've only just put a 7950X3D in the newest box and finally I no longer see any lags with an RTX 4090 at 4k. I guess that will be just fine for a while and by the time I'm ready to fall for a 9950X3D, it will be quite less than the original price.

I sure won't make the mistake of upping resolutions again, going beyond my beautiful Eizo 1920x1200 meant a decade of game performance misery, 4k has to be good enough for a long time.

 

[NinoPino]

No, intel cpus are vastly easier to cool. Vastly. They just have better heat dissipation. How do you think reviewers are able to hit 300 and 350w power draw out of these chips? It's because they are easy to cool

Very easy to cool, sure, and rock solid. We are seeing.

 

[usertests]

I'm not normally one to complain about such things, but it's so bad that it really affects their readability. So, um, like what gives, AMD??

Could be the same trash marketing team that gave incorrect info about previous launches, or it might be trickery to prevent or catch slide leakers.

 

[NinoPino]

Uhm, have you heard about high vsoc making amd cpus immolate and cook the mobo alongside them?

Very bad situation. AMD tried to hide something ? Or talked about a "not identified problem" ?

It wasn't too long ago.

Intel problem started long ago but it is still here.