Why everyone say Zen 5 is bad ?

[R0H1T]

And I think you answered why WIN 11 is bugged now, Arrow Lake has no Hyperthreading, so now the thread manger in windows needs to be updated, and I bet that is what interduce this bug.

Speculation, I haven't seen any reason as to why games or any other processes need elevated privileges for them to get more performance on AMD ~ If HuB's results are reproducible you have to question what the heck MS is doing with Windows these days

3-4% higher fps(min?) is no joke!

 

[Hyderz]

I haven't seen any reviews of 9000 series with windows 10 vs 11
my money is on windows 10 with the higher performance numbers since we are talking about microsoft bugs

 

[NSX05]

I kind of frustrated with this.

I get that for example 9700X is on par with performance of 7700X, however is it REALLY enough to claim Zen 5 is "DOA" or "Bad" ?
My point is, it's doing that with lower clock, since all core 9700X = 4480MHz vs. 7700X = 5190MHz, AND a lot lower power usage (9700X = 88W vs. 7700X = 148W).
Relevant tables from GN video :

I think it's more the forced false narrative that Zen 5 is much more efficient than Zen 4, which was a choice AMD made to focus on, and one chosen to the detriment of its performance.

Zen 5 consumes as much power as Zen 4 when properly configured, and vice versa. The 9700X is especially starved of power because it can't hit the frequencies that the 7700X can.

As a result it performs worse and costs more than a 14700K. And the story is very similar for the 9600X and 13600K.

It very much feels like AMD tried to repeat the stance that it had with mobile recently, allowing reviewers to test against AMD and Intel chips with higher TDP. That worked in AMD's favour there, but it hasn't here. Pricing is a whole other issue.

It simply leaves me recommending Zen 4 to people, and I don't think Zen 5 is bad in any way. It just ... exists as an option. The 9800X3D might be a different story when it arrives.

 

[Assimilator]

Really, what hurt AMD most is releasing new CPUs before the new 800-series boards are ready. Reviewers would have been a lot more forgiving of the lacklustre Zen 5 performance uplift if paired with a board with better features, i.e. new CPU + new board together would've be a more compelling launch than new CPU only.

... except for the fact that the only new thing on the 800-series boards is USB4, which was intended to be on the Zen 4/600-series boards at launch, but ASmedia has taken over 2 years to get right (a lot of which can be blamed on Intel's poor specs, but still). Feature stagnation (= PCIe lane counts) is what will kill AM5.

Unless we're seeing budget 3D models in the future, AMD would have nothing to compete with in the budget segment, even if that's not a first.

AMD's "strategy" for budget CPUs seems to be AM4, which is just plain stupid because it's prevented them from focusing all-in on AM5 and making that product line an attractive option across all price segments. Plus they have to maintain 2 microcode branches instead of 1, two sets of chipset drivers, two RMA channels, two production lines, ...

No Im saying that with TDP boost, more aggressive clocking, and working out of the bugs with windows fixes it will get to ~10%, add X3D to that... you've got yourself a nice chip.

AMD Fine Win strikes again! Don't buy the product that's reviewed, buy a hypothetical future product that may or may not ever perform at the levels claimed by a single AMD fanboy huffing copium!

Speculation, I haven't seen any reason as to why games or any other processes need elevated privileges for them to get more performance on AMD

It's not speculation, it's AMD fanboy copium.

 

[#22]

Quite nice piece of testing by Digital Foundry who seem to be back to CPU testing done their way - focusing on showing exact performance, so crucial consistency, not just averaged graphs leaving space for guessing:

https://www.eurogamer.net/digitalfoundry-2024-amd-ryzen-9-9900x-9950x-review?page=2

Know that sometimes these sites don't work and you see only tables with average fps, so not what I talk about:

Good example is comparing in 1080p Flight scoring similar mean averages 7800X3D and 14900K when first not much loosing in 1% lows. We can see that similar mean average comes from Ryzen providing often higher, but choppy framerate compared to smoother Intel. These tests also highlight what used to confuse me with averages scored by X3D chips in general. Sometimes - not in every game or even review - they tending to score suspiciously big distances between average fps and lows. Such suggest more stuttery performance, but you can only guess when seeing averaged graphs.

 

[Melvis]

I dont see Zen 5 as bad! just not as good as we are all so used to seeing from one gen to the next. It seems to really shine in Linux? from what im hearing which isnt a bad thing honestly as Linux will be growing more and more. Even Linus said it was good and coming from him thats a surprise when he is so Intel Bias!

 

[phanbuey]

Really, what hurt AMD most is releasing new CPUs before the new 800-series boards are ready. Reviewers would have been a lot more forgiving of the lacklustre Zen 5 performance uplift if paired with a board with better features, i.e. new CPU + new board together would've be a more compelling launch than new CPU only.

…

AMD Fine Win strikes again! Don't buy the product that's reviewed, buy a hypothetical future product that may or may not ever perform at the levels claimed by a single AMD fanboy huffing copium!


It's not speculation, it's AMD fanboy copium.

Don’t get me wrong - I love sh*tting on zen 5 release. Would I buy one - nope. I didn’t buy zen4 either - bad product.

but single core scores looks good and windows tweaks look good… X3D, esp on new boards, has a likely shot of being good.

 

[Shakallia]

I dont see Zen 5 as bad! just not as good as we are all so used to seeing from one gen to the next. It seems to really shine in Linux? from what im hearing which isnt a bad thing honestly as Linux will be growing more and more. Even Linus said it was good and coming from him thats a surprise when he is so Intel Bias!

Zen 5 is totally overpriced compared to its performance. For exemple, the 7700X is roughly behind by 3 %, but costs about 100 euros less than 9700X in Europe..So yea, on default a bit more efficient, but is it worth that price premium ? No.

There is absolutely no benefit into choosing those cpu's right now, unless for some very niche scenarios, unless AMD bring the price down, it will be a commercial disaster.

Not to mention Arrow Lake on the corner, if those deliver, AMD will be deep in it up to the head.

And plz nobody come saying that they are cheaper than ZEN 4 on launch, what matter is the price now...And ZEN 4 will give you 95+% performance for way cheaper, why bother really ?

 

[SL2]

I dont see Zen 5 as bad! just not as good as we are all so used to seeing from one gen to the next.

Sure, until you check out the price. Looking at performance without caring about price is usually pointless.

Will it go down? Yes, it always does.

I'm talking about launch price.

 

[phanbuey]

Sure, until you check out the price. Looking at performance without caring about price is usually pointless.

Will it go down? Yes, it always does.

I'm talking about launch price.

AMD Ryzen 7000 Zen4 CPUs are 5 times less popular than Zen3 according to December retailer sales - VideoCardz.com

^ everyone should be getting deja vu here.

 

[TheinsanegamerN]

I still see no reason to upgrade from my 5800x3d. Zen 5 doesnt really move the needle and I dont believe its x3ds will either. It makes major efficiency gains, so long as you dont look at 65w zen 4 parts.

The one thing its really good at is AVX512, which shows massive improvements. But we'll have to wait for more software that uses AVX 512.

 

[SL2]

everyone should be getting deja vu here.

AM5 board prices were higher back then, as well as DDR5, the were a LOT of people upgrading their existing AM4 machines, and 5800X3D was 7 months old and sold like crazy.
You could argue that the same goes for 7800X3D, but that one requires a new board for a lot of people.

If Zen4 didn't sell much in the beginning it was because of everything else than the CPU. Board, RAM, previous generation. At least the reviewers were much happier.

I didn't think Zen5 would ever attract people to upgrade from Zen4 in larger quantities anyway.

Being able to upgrade from a 2600 to a 5800X3D is a whole different thing and attracts a lot of buyers, AM5 doesn't offer that kind of leap.

 

[Zazigalka]

to me it looks like once you get a 7800x3d on pcie5 mobo like b650e, the whole upgrade path till am6 will be more like cosmetic changes. personally, I like it that way. Just give me the best performance for the foreseeable future up front, don't make me wait 2 gens for it.

Quite nice piece of testing by Digital Foundry who seem to be back to CPU testing done their way - focusing on showing exact performance, so crucial consistency, not just averaged graphs leaving space for guessing:

https://www.eurogamer.net/digitalfoundry-2024-amd-ryzen-9-9900x-9950x-review?page=2

Know that sometimes these sites don't work and you see only tables with average fps, so not what I talk about:


View attachment 359198



Good example is comparing in 1080p Flight scoring similar mean averages 7800X3D and 14900K when first not much loosing in 1% lows. We can see that similar mean average comes from Ryzen providing often higher, but choppy framerate compared to smoother Intel. These tests also highlight what used to confuse me with averages scored by X3D chips in general. Sometimes - not in every game or even review - they tending to score suspiciously big distances between average fps and lows. Such suggest more stuttery performance, but you can only guess when seeing averaged graphs.

wow, never seen cpu benches done this way, it's simply amazing. I checked out cp2077 1440p using 5800x3d vs 14600k, and x3d is better and more consistent, so I think the fps variance on x3d chips will, well, vary.... it wins in MSFS at 1440p too.

 

[igormp]

AMD Ryzen 7000 Zen4 CPUs are 5 times less popular than Zen3 according to December retailer sales - VideoCardz.com

^ everyone should be getting deja vu here.

Yeah, people complaining about an older product that already had lots of price cuts being cheaper than a new product don't really make much sense.
I don't think any gen-to-gen provided better value then its predecessor.
AMD's x3D models made this a bit weirder since now a non-X3D model will never provide the same value as a X3D one for gamers.

Being able to upgrade from a 2600 to a 5800X3D is a whole different thing and attracts a lot of buyers, AM5 doesn't offer that kind of leap.

...yet. Comparing a 7700x to a 9700x would be pretty similar to a 1700 to a 2700 back then, neither make much sense given the value.
For someone that's already on a first gen AM5 CPU, a worth upgrade will likely only become a thing by the end of AM5's lifespan, like the 5800x3D was.

That's only thinking about games, of course. For productivity it's a whole different thing, but then each person should know what their application requires in terms of horsepower, and how much value such uplift would bring (since less time spent processing stuff would mean more money, ideally).

 

[#22]

to me it looks like once you get a 7800x3d on pcie5 mobo like b650e, the whole upgrade path till am6 will be more like cosmetic changes. personally, I like it that way. Just give me the best performance for the foreseeable future up front, don't make me wait 2 gens for it.


wow, never seen cpu benches done this way, it's simply amazing. I checked out cp2077 1440p using 5800x3d vs 14600k, and x3d is better and more consistent, so I think the fps variance on x3d chips will, well, vary....

Results always depend on each test, so factors like platform, CPU settings, games or test location in game. The same with tendencies - just show stuff supporting any. I don't see a difference you mentioned in Cyberpunk, both 5800X3D and 14600K behave similarly, but maybe you clicked something wrong or just see things you want Btw suggest with numbers coming from lowest resolution - really no point looking at higher ones Or now check e.g. Flight 1080p. Here 14600K is rather undoubtly more consistent than 5800X3D, but maybe things look different in other games.

About you never seeing such tests, it's not popular and maybe you are new to hardware - Digital Foundry started CPU testing like that ten years ago and these guys are rather well known among at least gaming focused part of hardware world. They just stopped few years ago, around Intels 10th gen, and now it seems they come back.

 

[igormp]

It seems to really shine in Linux? from what im hearing which isnt a bad thing honestly as Linux will be growing more and more.

Phoronix released some benchmarks comparing Windows vs Linux with the 9950x (heads up that it has no games included):

Windows 11 vs. Ubuntu 24.04 Linux Performance For The AMD Ryzen 9 9590X Review - Phoronix

www.phoronix.com

TLDR; going from a 7950x to a 9950x on Linux provided a ~14% uplift, while on Windows it was a ~10% uplift. Using a 9950x, moving to linux from windows yields a ~11% perf increase.

It's funny how in some tests the 7950x on Linux manages to be faster than a 9950x on windows lol

 

[chrcoluk]

It's puzzling why AMD opted for the 9700X to be a 65W part, instead of ... say ... 80W: performance "@ stock" would be better and the efficiency wouldn't suffer as much, but IT WOULD make the gains over the 7700X "more palpable", instead of "rather small".

Some people want PCs to become more lower power again, and ultimately it is a better product at lower TDP in my opinion, however the price is bad, and even worse as pointed out in this new video, it doesnt even come with a cooler like the 7700 does.

But I think what has been overwhelming in the response to the reviews is most people seem to not care much about efficiency, they just want performance, which is interesting considering all the stuff being thrown in intels direction regarding high wattage.

If the story is true about the TDP being raised in a future agesa update then AMD must have been thinking, people are like cats, they not sure what they want.

 

[mkppo]

Results are all over the place, but as more benchmarks surface, looks like some of the applications I run (decryption and some machine learning apps) run about 20-30% faster than zen 4 even without avx512 . Here's hoping all the negativity and lack of sales translate into great deals over the coming months

 

[AusWolf]

Zen 5 is totally overpriced compared to its performance. For exemple, the 7700X is roughly behind by 3 %, but costs about 100 euros less than 9700X in Europe..So yea, on default a bit more efficient, but is it worth that price premium ? No.

If you compare it to the 7700 non-X, the situation is even worse. The 9700X only has better AVX-512, nothing else. It won't be a bad CPU once Zen 4 inventory clears, but until then, the 7700 is a much better buy.

AMD Ryzen 7000 Zen4 CPUs are 5 times less popular than Zen3 according to December retailer sales - VideoCardz.com

^ everyone should be getting deja vu here.

As long as they keep selling Zen 3, there's no surprise here. The average user doesn't need the extra CPU performance, especially when you combine it with the cost of the platform swap.

Edit: I see the article is from December. I'm wondering how much the situation has changed considering that AM5/DDR5 got a bit cheaper since then.

Some people want PCs to become more lower power again, and ultimately it is a better product at lower TDP in my opinion, however the price is bad, and even worse as pointed out in this new video, it doesnt even come with a cooler like the 7700 does.

But I think what has been overwhelming in the response to the reviews is most people seem to not care much about efficiency, they just want performance, which is interesting considering all the stuff being thrown in intels direction regarding high wattage.

If the story is true about the TDP being raised in a future agesa update then AMD must have been thinking, people are like cats, they not sure what they want.

It's not that people don't care about efficiency. It's that the 9700X has nothing to offer over the 7700. Same efficiency, similar performance. Boring.

 

[R-T-B]

when can't restrict poor ppl out

TIL poor people are not valid for statistics.

 

[Speedyblupi]

Yeah, people complaining about an older product that already had lots of price cuts being cheaper than a new product don't really make much sense.
I don't think any gen-to-gen provided better value then its predecessor.

Intel 12th gen did (i5-12600K was faster and cheaper than i9-11900K).
Ryzen 3000-series did (Ryzen 5 3600X was faster and cheaper than Ryzen 7 2700X).
We shouldn't expect to get these huge uplifts every generation, but we still hope for them, and even when the value of the new generation isn't better than the previous generation (as is usually the case) it's usually not as bad as Zen 5's launch prices.
To be fair, Zen 5's value proposition is still a bit better than 14th gen Core was, but several recent generations (notably 13th gen Core and Zen 3, which weren't cheaper on launch than their predecessors for the same level of performance, but were still a decent performance uplift per-tier, especially for gaming) have been far more compelling value options than either Zen 5 or 14th gen Core were at launch.

 

[oxrufiioxo]

TIL poor people are not valid for statistics.

It's the quad core 8 thread and 570/580/6600/1050ti/1060/2060/3060 crowd that dictates game development not the X3D/i9/90 series owners.... Somebody should probably tell him.

The budget offerings matter a lot more than the halo products and it's been stagnation for almost a half decade in that class of products.

Also most people still use 1080p and most pc gamers are playing Minecraft, Fortnight, the Sims, counterstrike, roblox.... games anything modern will run in it's sleep at least at the settings most use in them.

The 9600X being disappointing is the biggest blow becuase it is more likely to dictate future game development than the 9950X3D/9800X3D....

Even game consoles could end up with the Zen 5 core as their base and while that would be a nice uplift over the Zen 2 based cores they have now it'll still mean the general perfomance will be very outdated whenever they launch. All speculation of course but the low end and consoles still affect the upper range products in the long run.

 

[mechtech]

AMD Ryzen 7000 Zen4 CPUs are 5 times less popular than Zen3 according to December retailer sales - VideoCardz.com

^ everyone should be getting deja vu here.

What's surprising about that? DDR5 was initially quite a bit more $$ than ddr4, also, if you bought a zen 5k series, why would you upgrade to 7k so soon?? I don't put together a new PC until 5-6 years.

 

[oxrufiioxo]

What's surprising about that? DDR5 was initially quite a bit more $$ than ddr4, also, if you bought a zen 5k series, why would you upgrade to 7k so soon?? I don't put together a new PC until 5-6 years.

Yeah and my 7950X3D obliterates my 5950X so if the Zen4 uplift didn't excite people this definitely won't.

 

[Athlonite]

Steve from HUB just told everyone in a new video that according to AMD if want to get the same results as us you need to benchmark it's Zen 5 CPU's using the hidden System Administrator account if not you're going to see lower results when using the standard User Administrator account

You can test this for yourself if you like by opening a CMD prompt and typing the following

net.exe user administrator /active : yes

then log out and log in as the system admin... But be warned doing this can leave you extra vulnerable to malware as you'll now be running everything with admin privileges