AMD Ryzen 9 9950X

[oxrufiioxo]

To be fair, at least AMD doesn't require a new socket with every new generation.

We will see, if zen6 requires a new socket or it also is a meh generational improvements does it even matter how long their marketing department says they are supporting a socket.

Balls in AMDs court on this one but if this ends up the last hurrah of AM5 and all we get is some meh apu next year and it's done this socket was overall not any better than what Intel has been doing forever.

 

[rv8000]

Interesting take consideration I haven't been a fan of 13th or 14th generation mostly due to how they are configured out of the box and the E cores. I liked 12th generation though and 13th offered more uplift over 12th than this and is better at gaming even though it's 2 years old. Intel offering meh performance increases isn't a excuse for AMD offering almost 0 uplift over a 2 year old architecture.

The but Intel did this or that really doesn't matter I only expect these to be mildly impressive over 2 year old cpus. Assuming this generation last another 2 years going nearly a half decade at a similar perfomance level isn't good for anyone it wasn't good when Intel did it from 2012-2017 it isn't good now.

I've owned at least 2 ryzen cpus from each of the last 3 generations and 1 from 2000 so Intel really hasn't been on my radar whatsoever I do have a decent amount of hands on time with i7/i9s from the previous 3 generations though.

I agree that although they are terrible when it comes to generational improvements pricing is the biggest issue currently these make the 4060ti 8G blush when it comes to P/P uplifts. AMD also better hope Arrow Lake is a dud it doesn't have to do much over raptorlake to be significantly more impressive.

It’s not really. Your post here is in pretty stark contrast for the 9950X comparatively. It’s interesting you gave the 14900k a pass, an entire paragraph how you’re hopeful for it and APO, and saving overall the performance is good. Meanwhile “Wow! This is bad.” for the 9950X.

But anyways…

 

[dir_d]

I wanted to wait for this review before i commented on this new arch. Its clearly made for Servers/Workstations and there is 0 benefit to upgrade from AM4 or the Zen 7000 series to these chips. Now i look forward to see what Intel can bring us.

 

[Vincero]

We will see, if zen6 requires a new socket or it also is a meh generational improvements does it even matter how long their marketing department says they are supporting a socket.

Balls in AMDs court on this one but if this ends up the last hurrah of AM5 and all we get is some meh apu next year and it's done this socket was overall not any better than what Intel has been doing forever.

True, but Intel plans that - purposely so and not even with sockets but also chipsets and CPU support... as has been demonstrated by those who modded LGA 1151 boards to work with CPUs from every iteration.

In reality though, unless they are going to completely re-layout the pin assignment for the socket, I can't see what benefit there would be - it's not like older sockets going to north/south bridge chipset designs that could be impacted by that chipset performance. With the IMC, core logic, PCIe controller, all on the chip package if there is a performance issue it should be fixable changing those package components - unless there is some real electrical limitations otherwise that shouldn't need a new pin-out.... BIOS support though could be a bigger issue - a repeat of the AM4 Zen1>2 only or 2>3 only BIOS support limitations would not be fun.

 

[BoggledBeagle]

Now i look forward to see what Intel can bring us.

a chip mosaic

 

[oxrufiioxo]

It’s not really. Your post here is in pretty stark contrast for the 9950X comparatively. It’s interesting you gave the 14900k a pass, an entire paragraph how you’re hopeful for it and APO, and saving overall the performance is good. Meanwhile “Wow! This is bad.” for the 9950X.

But anyways…

I mean it's a refresh of a 2 year old architecture and still perfoms better in games and similar in applications.... I'm not a fan of it so why would I like this?... the 13900k was fine if it actually worked properly out of the box it still offered more than this even though intel typically doesnt offer in socket cpu that are worth upgrading from the first generation. The 12900k is fine considering it's 2.5 years old and launched when its only competition was ryzen 5000....

Intel has stagnated post 13th gen and sorta since 12th generation now amd is doing the same although who's doing it worse is debatable due to intel having a new architecture out this year and raptorlake being 2 years old. Like I said arrowlake doesn't even have to be all that impressive to be significantly more impressive than this AMD really set the bar low for intel.

 

[KaitouX]

It's more than just parking issues, linux shows 17.8% uplift across some ~400 odd tests!

AMD Ryzen 9 9950X & Ryzen 9 9900X Deliver Excellent Linux Performance Review - Phoronix

www.phoronix.com

Looking at individual results of things tested by many outlets seems to show that the performance is about the same on Windows and Linux. The biggest difference seems to come from testing a lot of things that benefit from the improvements made to Zen5.
That's not to say that there aren't issues with Windows, as they exist, but I wouldn't expect the average performance in common workloads to change that much, probably it would be closer to the difference between the 9700X and 7700.

It’s not really. Your post here is in pretty stark contrast for the 9950X comparatively. It’s interesting you gave the 14900k a pass, an entire paragraph how you’re hopeful for it and APO, and saving overall the performance is good. Meanwhile “Wow! This is bad.” for the 9950X.

But anyways…

I mean, the 14900K is a refresh 1 year after the 13900K. This is a new arch 2 years later. Obviously the expectations were higher, and so the reaction to the small improvements in most applications is also going to be stronger.

 

[Vincero]

a chip mosaic

Hahaha, it wasn't until I read that I thought about this 'artwork'...

Back to business, Intel's foveros packaging solution may offer some benefits... it's what's in that package that's important.... or rather how much TDP control there is of all those tiles and the package in general... they can do the same as AMD and keep some tiles the same and sub in new ones so long as they fit.

 

[AnotherReader]

Looking at individual results of things tested by many outlets seems to show that the performance is about the same on Windows and Linux. The biggest difference seems to come from testing a lot of things that benefit from the improvements made to Zen5.
That's not to say that there aren't issues with Windows, as they exist, but I wouldn't expect the average performance in common workloads to change that much, probably it would be closer to the difference between the 9700X and 7700.


I mean, the 14900K is a refresh 1 year after the 13900K. This is a new arch 2 years later. Obviously the expectations were higher, and so the reaction to the small improvements in most applications is also going to be stronger.

At least for code compilation, Linux benchmarks are far better for Zen 5.

Compare this to TPU's code compilation test.

 

[BoggledBeagle]

Back to business, Intel's foveros packaging solution may offer some benefits... it's what's in that package that's important....

I looked in my crystal ball and saw Intel engineers pulling their hair out, cursing badly and wishing they stayed with a monolithic design...

Let us hope their artpieces will run as intended and will not fall apart.

 

[oxrufiioxo]

At least for code compilation, Linux benchmarks are far better for Zen 5.

View attachment 358996

Compare this to TPU's code compilation test.

View attachment 358997

These are 100% server focused cpus I'm just worried if that's the route amd takes going forward general desktop/gaming performance will suffer. I mean these have regressed even depending on how you test them and a new cpu at a minimum shouldn't lose to and old one getting ptsd from the 11900k launch smh the generation that made me skip intel for going on a half decade now.

We got large generational gains from 2000-3000, 3000-5000, and 5000-7000 so getting what amounts to negligible gains this generation is disappointing and I'll give amd a pass if Zen6 is both good and on AM5.

I looked in my crystal ball and saw Intel engineers pulling their hair out, cursing badly and wishing they stayed with a monolithic design...

Let us hope their artpieces will run as intended and will not fall apart.

Yeah that's the X factor for sure meteorlake was a bust on desktop and at least to me isn't that impressive on laptops although I'm sure their are scenarios where it makes sense.

So Intel still needs to prove the tile based solution will work on Desktop we got a couple months before we find out.

 

[R0H1T]

Nah. Afaik, optical interconnects are still not ready for prime time. Also, i doubt anything outside Data Centers will make use of them.

I was thinking about TSMC's integrated fan-out. There are other options, but those are probably too expensive for Ryzen.

Outside of 3d stacking I don't really know how you can solve these without massive latency issues?

Analysis of CPU performance counters in video transcoding and code compilation indicates that Zen 5 suffers more from memory latency than Zen 4. I expect the uplift from 3D cache to be higher than for Zen 4. Whether it will be enough to be significant is another matter altogether.

zen5 said ouch

Core to core latency is bad as well, think of it how the internet could be choked by just a handful of tubes cables

Interesting take. It’s okay when Intel offers a smaller performance increase at more power, but bad when AMD offers the same performance increase at less power. Not to mention non consumer oriented workloads see a healthy benefit on top on some more regular single threaded applications.

It's not just about Intel vs AMD, whether MS has botched win11 wrt zen5 or AMD didn't work with them long enough to optimize performance on Windows the fact remains it's severely underperforming on the most popular desktop platform. At the end of the day plebs won't care who screwed them as AMD should be carrying the burden of their products!

These are 100% server focused cpus I'm just worried if that's the route amd takes going forward general desktop/gaming performance will suffer.

They're definitely not 100% server focused, otherwise AMD wouldn't bother selling them at half or a quarter of their server margins!

It would be interesting to see if better memory speed/timings would have any (major) impact here. Or maybe a magical AGESA fix just for Windows

 

[oxrufiioxo]

They're definitely not 100% server focused, otherwise AMD wouldn't bother selling them at half or a quarter of their server margins!

bad wording I meant the actual generational improvements are aimed at servers.

These are still just fine for desktop use it's not like Ryzen 7000 is bad even by today's standards. This is just disappointing and overpriced.

 

[rv8000]

Looking at individual results of things tested by many outlets seems to show that the performance is about the same on Windows and Linux. The biggest difference seems to come from testing a lot of things that benefit from the improvements made to Zen5.
That's not to say that there aren't issues with Windows, as they exist, but I wouldn't expect the average performance in common workloads to change that much, probably it would be closer to the difference between the 9700X and 7700.


I mean, the 14900K is a refresh 1 year after the 13900K. This is a new arch 2 years later. Obviously the expectations were higher, and so the reaction to the small improvements in most applications is also going to be stronger.

While from a gaming standpoint there’s no reason to applaud Zen5, productivity workloads and more prosumer workloads that were not tested in the TPU test suite provide gains anywhere from 10-25% over the 7950X.

It sucks the wider, AVX 512 oriented design didn’t provide massive gains in every workload but it’s nothing like the lack of improvements (non existent excluding pumping TDP to push clocks) going from 13th to 14th gen. Hence my comment on his totally moronic statement (as per usual).

They’re good CPUs. AMD just shot themselves in the foot value wise by letting previous gen chips slide tremendously down in price. As for marketing, people really need to stop expecting anything until 3rd party reviews hit, we only have ourselves to blame if we go along for any tech companies ride.

 

[Event Horizon]

A flop is a flop. Prices need to come down, and marketing departments need to return to reality.

 

[AnotherReader]

These are 100% server focused cpus I'm just worried if that's the route amd takes going forward general desktop/gaming performance will suffer. I mean these have regressed even depending on how you test them and a new cpu at a minimum shouldn't lose to and old one getting ptsd from the 11900k launch smh the generation that made me skip intel for going on a half decade now.

We got large generational gains from 2000-3000, 3000-5000, and 5000-7000 so getting what amounts to negligible gains this generation is disappointing and I'll give amd a pass if Zen6 is both good and on AM5.



Yeah that's the X factor for sure meteorlake was a bust on desktop and at least to me isn't that impressive on laptops although I'm sure their are scenarios where it makes sense.

So Intel still needs to prove the tile based solution will work on Desktop we got a couple months before we find out.

Code compilation and web browsers see significant uplifts and the latter is very relevant to the average user especially with the use of Electron for desktop applications. I haven't seen any analysis of games, but I suspect it's primarily due to games having higher L3 miss rates than most applications.

 

[Carillon]

This is a new arch 2 years later. Obviously the expectations were higher, and so the reaction to the small improvements in most applications is also going to be stronger.

The expextations were higher because AMD set them that high. It would have been fine if they said zen5 was for professionals, but they instead claimed it to be the best gaming CPU ever.

 

[R0H1T]

We're splitting hairs as usual, marketing even at its best is generally deceptive (up to?) & exaggeration is common place!

The best way to deal with most product launches is not to expect anything, of course I'd prefer if most all of them were free

 

[oxrufiioxo]

We're splitting hairs as usual, marketing even at its best is generally deceptive (up to?) & exaggeration is common place!

Still feels like I'm in the twilight zone the same people who bashed the 13900k pre degradation issues being a known issue are now championing this cpu as our lord and savior smh....

The marketing was more deceptive than usual for AMD almost rdna3 levels of bad but even when I pointed out the marketing didn't make sense AMD fanboys still took it as gospel and are now defending this pile of shite (once the price drops like a rock they'll be fine though)

I can't remember how many people told me the gaming uplift was good for 2 years post zen4 and that it was much faster than raptorlake cuz amd said so...

 

[Paganstomp]

Well, that landed with a spectacular thud. Guess I'll be staying with my AM4 B450 system for a few years more. Considering I just got a Ryzen 9 5900x. I'm good.

 

[DemonicRyzen666]

Still feels like I'm in the twilight zone the same people who bashed the 13900k are now championing this cpu as our lord and savior smh....

Meanwhile, the 14900K going from the 13900K gained anywhere from 3.1% up to 5.6%. in IPC a slight power decrease.

 

[Ruru]

So for users already invested in AM4(Ryzen 5000), still no reason to upgrade.

At least when gaming at 4K.

 

[oxrufiioxo]

Meanwhile, the 14900K going from the 13900K gained anywhere from 3.1% up to 5.6%. in IPC a slight power decrease.

Correct but came out much closer together and was stated to be a refresh by intel from the start. And is equally meh.

 

[R0H1T]

Still feels like I'm in the twilight zone the same people who bashed the 13900k pre degradation issues being a known issue are now championing this cpu as our lord and savior smh....

The marketing was more deceptive than usual for AMD almost rdna3 levels of bad but even when I pointed out the marketing didn't make sense AMD fanboys still took it as gospel and are now defending this pile of shite (once the price drops like a rock they'll be fine though)

I can't remember how many people told me the gaming uplift was good for 2 years post zen4 and that it was much faster than raptorlake cuz amd said so...

The issue with Intel, for me, back then wasn't anything related to stability or whatever else. It was the absurd clocks, the chips are already pushed way beyond sane limits at "13th" gen, I also said AMD was stupid to follow Intel down that rabbit hole but at least so far as winning benchmarks was concerned it was "somewhat" understandable. Though personally I wouldn't buy any of 7900/50x & let it go untethered at those clocks/power limits! Intel's created a massive headache for themselves as well because for the first time in a long long time we will probably see a major(?) clock regression coming from 14th gen to whatever they release now.

 

[tfp]

Thank you for the article @W1zzard.

One general comment on the graphs, it would be great if the processor(s) we should compare with where in bolder colors. The 7950X was in a light blue gray that is, at least for me, hard to see and it would be great if the 14900 was called out as well in it's own color. I was able to find them but it would be nice to see them at a glance.