Why everyone say Zen 5 is bad ?

[agent_x007]

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 :

What is wrong with everyone dumping on CPU that is clearly better than previous one ?
This is similar situation to Core 2 Duo E6300 vs. Pentium Extreme Edition 965 (just not as extreme, since we have very different pricing, but that's just future for you).
Everyone clearly knows which one is better of the two, even if both have similar performance :

So, WTF reviewers ?
From what I see, the only bad part about those new AMD CPUs is price, but that will be adjusted later (as always).
Also, I would love to see overclocking performance and power increase associated with it, however (I guess ?) early BIOSes/AGESA aren't stable enough for it ?

Lastly, I'm really afraid of everyone always expecting performance jumps of 20-25% between generation, when frequency scaling is TOUGH on higher end of the scale. Intel clearly shown where limits of that scale lay (both now, and in Pentium 4 days), and what are consequences of pushing blindly for frequency increases. Seeing frequency regression is really good, when paired with similar performance vs. other stuff.
We really don't need more frequency wars (neither on CPUs, nor on GPUs), and there is more to good CPU than just performance vs. previous gen.

 

[tabascosauz]

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 :
View attachment 357853
View attachment 357854

What is wrong with everyone dumping on CPU that is clearly better than previous one ?
This is similar situation to Core 2 Duo E6300 vs. Pentium Extreme Edition 965 (just not as extreme, since we have very different pricing, but that's just future for you).
Everyone clearly knows which one is better of the two, even if both have similar performance :

So, WTF reviewers ?
From what I see, the only bad part about those new AMD CPUs is price, but that will be adjusted later (as always).
Also, I would love to see overclocking performance and power increase associated with it, however (I guess ?) early BIOSes/AGESA aren't stable enough for it ?

Lastly, I'm really afraid of everyone always expecting performance jumps of 20-25% between generation, when frequency scaling is TOUGH on higher end of the scale. Intel clearly shown where limits of that scale lay (both now, and in Pentium 4 days), and what are consequences of pushing blindly for frequency increases. Seeing frequency regression is really good, when paired with similar performance vs. other stuff.
We really don't need more frequency wars (neither on CPUs, nor on GPUs), and there is more to good CPU than just performance vs. previous gen.

That's not so much credit to 9700X for being efficient as it is a knock on the 7700X for being what it is - a rushed knee-jerk reaction to try and push more performance than was efficiently available from the node at the time (and retroactively trying to paint it in a better light with the 95C justifications). AMD only released Zen 4 parts with more reasonable power limits afterwards. When N5 Zen 4 isn't trying to reach utterly insane Fmax and staying in its 5GHz and under comfort zone, the temps and power aren't much different.

I don't know who's using the labels "DOA" and "bad", but it's not a particularly exciting product either. Ryzen on N4 is not really anything we haven't seen already, it's why Phoenix, Hawk, and Strix efficiency are so good. Add to that the fact non-X3D 1CCD is still sitting in an awkward place, with no real changes in price to excite people.

And now the freq gap between X3D and non X3D seems to be closing, not with the X3D moving up but with the regular parts moving *down* in Fmax, 9700X is in a bit of a weird place.

 

[dirtyferret]

it's the problem when you cherry pick marketing result to show people with 10-20% gains in Far Cry 6 and LOL, the reality is those gains are outliers and in some games the 7700x performs better than the 9700x.
Is it a bad CPU? No, is it a currently good CPU for the price? No. Are people pissed AMD over hyped the CPU, I'm sure they are.

 

[AnotherReader]

it's the problem when you cherry pick marketing result to show people with 10-20% gains in Far Cry 6 and LOL, the reality is those gains are outliers and in some games the 7700x performs better than the 9700x.
Is it a bad CPU? No, is it a currently good CPU for the price? No. Are people pissed AMD over hyped the CPU, I'm sure they are.

Marketing pulled a fast one on us by focusing on IPC. According to Gamers Nexus's review, clocks might be lower than Zen 4 in all core workloads. An IPC uplift with reduced clocks is in line with the results that we see.

 

[JohH]

It's the typical pipeline of expectations and disappointment.
Expectation: AMD said these were gaming leadership parts.
Reality: They are not.

Why are people disappointed?
1. AMD released the low TDP parts first.
2. AMD set the price a bit high.
3. AMD still hasn't increased core count since 2019.
4. AMD hasn't been able to launch X3D parts with the initial parts - if this continues happening they will probably always be disappointing and hurt their products initial impressions.
5. Not AMD's fault but there were rumors of even more insane results. While these rumors should have been discarded as false, some people still believed them.

 

[natr0n]

They are stable cpu unlike intels mess right now which is good.

 

[#22]

To OP, performance progress made going from previous models is negligible. Especially when we compare it with uplifts of previous gens of Zen. Zen 5 is more similar to what Intel used to do in previous decade. And what will be with next gen X3D, similarly? Many people may also start to feel bad in terms of mythical and theoretical "uprade path" - let's assume another gen bringing similar uplift and it being it for AM5. Option to upgrade for CPU not worth upgrading, but all of these gens of compatible parts flooding the market to massacre resell value of your already owned chip and mobo. Potentially more than cost of next motherboard. For now it looks worse than "upgrade path" of 1700 already serving as example for a while.

Nicer power numbers come mostly from setting default behaviour of CPU this time more sensibly - search for some power normalized benchmarks comparing both AM5 Zens.

 

[Solaris17]

however (I guess ?) early BIOSes/AGESA aren't stable enough for it ?

I mean idk about anyone else but before I even judge performance I generally just want my PC to start.

 

[Carillon]

As somebody mentioned in the other thread, the 9700x has the same TDP as the 7700 non-X. And it's not much better!
If you limit the 7700x to the same TDP it will have the same efficiency as the 7700 non-X.
All the efficiency improvements of the 9700x vs the 7700x are just because it's sitting on a more sane point on the voltage/frequency curve, and the small node advancement.

 

[Dr. Dro]

It's because it doesn't leave Raptor Lake in the dust and the gains over Zen 4 tend to range from non-existent to minor. People are way too caught up with raw performance numbers and seem to forget what's imo most important - this little chip performs amazing while keeping the watts very low. It gives me Core 2 vibes. I'm very impressed.

 

[AnotherReader]

It's because it doesn't leave Raptor Lake in the dust and the gains over Zen 4 tend to range from non-existent to minor. People are way too caught up with raw performance numbers and seem to forget what's imo most important - this little chip performs amazing while keeping the watts very low. It gives me Core 2 vibes. I'm very impressed.

Judging by some of the tests in Phoronix's review, life for ARM alternatives and Intel's Xeons is about to get even harder.

 

[Frick]

it's the problem when you cherry pick marketing result to show people with 10-20% gains in Far Cry 6 and LOL, the reality is those gains are outliers and in some games the 7700x performs better than the 9700x.
Is it a bad CPU? No, is it a currently good CPU for the price? No. Are people pissed AMD over hyped the CPU, I'm sure they are.

To be fair it says "up to" right there.

 

[agent_x007]

I don't know who's using the labels "DOA" and "bad", but it's not a particularly exciting product either.

HUB review ("Flop" thumbnail one).

It's because it doesn't leave Raptor Lake in the dust and the gains over Zen 4 tend to range from non-existent to minor. People are way too caught up with raw performance numbers and seem to forget what's imo most important - this little chip performs amazing while keeping the watts very low. It gives me Core 2 vibes. I'm very impressed.

I agree. Especially now after I saw Derbauer's video with actual OC numbers :

 

[Dr. Dro]

HUB review ("Flop" thumbnail one).


I agree, especially now after I saw Derbauer's video with actual OC :

Yep, I took the time to bench after someone linked me that in the review thread. I'll just drop the link here if you're interested. Still a minor loss vs. a 13900KS/14900K's P-cores at default settings in R23 even after giving it twice the wattage, which leads me to believe this 65W TDP/88W PPT spec is the sweet spot for this chip. This much performance at only 80 watts is really a grand achievement, people are not counting that as much as I would like to see.

 

[Bobaganoosh]

It really all depends on what you want to do with it.

If you want to play games with it, there's no incentive in any of the testing that the price is justified in comparison to a 7800X3D. Even from an efficiency standpoint, the 7800X3D wins.

Further, this response is 100% why we are in the boat with Intel right now. We mostly care about performance, and as gamers make up the main part of the consumer segment for these parts, they are the ones that look generation to generation and have decided that performance uplift gen-over-gen is the most important thing. So being that the gaming performance is roughly on-par with 7700X, that's being considered a failure by many. They're doing it at lower clocks and much lower power, which is awesome, but they're doing worse than an older part that's the same price or cheaper and they're not beating the part it's replacing. Intel looked at this type of response and couldn't compete at the same power levels, so they've just been cranking their products up to 11 and now we're seeing 13th and 14th gen failures because they were complacent about it and allowed voltages to creep too high. If they just released parts at the same or lower power every year with just IPC gains, Intel wouldn't be competitive, but they know that gamers care more about performance increase than power draw and they made that gamble.

GPUs are in the same boat and melted connectors is just one of the downsides we've seen and will see if they keep increasing the power draw for these.

 

[ARF]

The problem is that Apple has just started to use the state-of-the-art TSMC 2nm process, while AMD (5nm to 4nm is only a plus, 5+), intel and nvidia use ancient processes, extremely emphasized by the intel failures.
If they continue to avoid making technological progress with the lithography, things would become only worse in the next years.

We need more cores now, not some weird new instructions and hardware layout that is bet to be proof sometime in the distant future, and not now.

 

[agent_x007]

I always wanted to see a throwback to OK-ish stock performance, and decent/great overclock headroom on CPUs (good ol' Core 2 stuff).
Maybe we are getting a glimpse of that with 9700X ?
We shall see...

The problem is that Apple has just started to use the state-of-the-art TSMC 2nm process, while AMD, intel and nvidia use ancient processes, extremely emphasized by the intel failures.
If they continue to avoid making technological progress with the lithography, things would become only worse in the next years.

We need more cores now, not some weird new instructions and hardware layout that is bet to be proof sometime in the distant future, and not now.

When Apple makes x86 core, we can have a conversation about it. Right now, that's 100% pointless/irrelevant to current situation.

 

[ARF]

When Apple makes x86 core, we can have a conversation about it. Right now, that's 100% pointless/irrelevant to current situation.

I didn't expect so silly response, but will answer your question.

Why everyone say Zen 5 is bad ?​

Because it lacks cores. Because it is made on the ancient 5nm process, and people feel there is no initiative to buy.

 

[Dr. Dro]

I always wanted to see a throwback to OK-ish stock performance, and decent/great overclock headroom on CPUs (good ol' Core 2 stuff).

That's the vibe I got from this launch. Sans the free +100% OC on the dinky stock cooler you got out of Conroe, that is.

Because it lacks cores. Because it is made on the ancient 5nm process, and people feel there is no initiative to buy.

Brother, you're trying too hard. Since when is TSMC N4 an ancient process? Please lol

 

[ARF]

Brother, you're trying too hard. Since when is TSMC N4 an ancient process? Please lol

Since there are two node shrinks afterwards, if you don't know about them. TSMC 3nm and TSMC 2nm.

AMD just showed us the same old - to snatch defeat from the jaws of victory.

 

[Onasi]

The problem is that Apple has just started to use the state-of-the-art TSMC 2nm process, while AMD (5nm to 4nm is only a plus, 5+), intel and nvidia use ancient processes, extremely emphasized by the intel failures.
If they continue to avoid making technological progress with the lithography, things would become only worse in the next years.

When Apple makes x86 core, we can have a conversation about it. Right now, that's 100% pointless/irrelevant to current situation.

More importantly, there is a very good reason why new bleeding edge processes usually start out with Apple and their low-power SOCs - initially these processes are just not a good fit for power-hungry desktop parts and the yields would be abysmal. Apple is essentially both a test run and a stabilizer. Even if NV or AMD COULD buy some 2nm allocation there is no way they would WANT to. I think ARF thinks that new nodes are magic that just by itself makes any chip design better and works OOB. It's not.

 

[agent_x007]

I didn't expect so silly response, but will answer your question.

Why everyone say Zen 5 is bad ?​

Because it lacks cores. Because it is made on the ancient 5nm process, and people feel there is no initiative to buy.

So... you think everything below 12 core on Zen 5 is bad, because multicore performance isn't where it should be... OK.
Then, by your definition, AMD should stop making everything below Ryzen 9 - since "they don't have enough cores".
...
Good luck with that.

 

[Dr. Dro]

Since there are two node shrinks afterwards, if you don't know about them. TSMC 3nm and TSMC 2nm.

AMD just showed us the same old - to snatch defeat from the jaws of victory.

Big deal. They're either fully booked years in advance by customers with bottomless pockets (Apple) or at an experimental stage, clearly unable to meet performance characteristics required to build the chip. Even Intel's 10 nm (Intel 7) is an extremely advanced node. It just doesn't work that way... besides, you want this thing to cost $3799 instead of $379, have the same or worse performance, and simultaneously be perma out of stock?

I fed you enough man, not cool to troll up a fair topic like this one

 

[freeagent]

For guys like me on Zen 3 it’s a solid upgrade. Zen 3 to Zen 4 was a bit to meh for me. I would imagine that would be how some guys feel.

 

[Darkholm]

I didn't expect so silly response, but will answer your question.

Why everyone say Zen 5 is bad ?​

Because it lacks cores. Because it is made on the ancient 5nm process, and people feel there is no initiative to buy.

Whatever you smoking, please send me some.