Review AMD Ryzen 5 9600X and Ryzen 7 9700X Review: Zen 5 brings stellar gaming performance

[-Fran-]

Guys. Take a break and go eat something, LOL.

Regards.

 

[bit_user]

When im comparing MSRP you tell me MSRP is useless.

There's no contradiction, here. If you were comparing Intel Gen-on-Gen MSRP, then I wouldn't complain. There is a direct correspondence between each of those products and it's informative to look at what the manufacturer does with pricing of their product tiers.

You're trying to manufacture a controversy where there really is none.

Anyways, even at MSRP, the 7700 is cheaper, since it launched at 330$.

As you're no doubt aware, AMD doesn't regard the 9700X as a direct replacement for the 7700. We'll have to wait and see why not, but maybe they have something planned for the 9700.

Really? A 22% increase in power draw for a 16% increase in performance is indeed something you don't see every day, and it's something you want to see every day cause it's horrible. The 9700x consume 22% more power than the 7700 in ST workloads.

Spoiler

Okay, so 22 W vs. 18 W is worth understanding. We'll have to see if this discrepancy holds with newer BIOS. In any case, it's definitely an improvement vs. the 24 W used by the 7700X.

Bud, im talking about the 7700. You know, the equivalent 65w cpu. It was 330$

Yes, some reviewers are comparing to the 7700, on that basis. I think that's an interesting comparison, but that the 7700X is the more logical point of reference. Even in that case, the 9700X offers a 12.5% improvement in single-threaded cinebench scores.

 

[stuff and nonesense]

Are we corporate clients here? Come on now. And I don't think corps would be buying this product anyways.

They won’t today, it’s too new, but if it is appealing they will test, validate and evaluate. Then they may or may not.

 

[YSCCC]

If you buy on the bleeding edge, 9700x you expect to tinker, update bios etc… for now same with intel when their next new kit comes out. Corporates wait a while, test, validate and buy kit with a long term supply guarantee so that 18 months down the line they can get an identical model. They don’t want exceptions.

Next replacement cycle, rinse and repeat…

For me anyone buying the bleeding edge isn't the target for the 9700X and 9600X... they are the latest gen, or a refresh of AM5 cpus, and the lower tier ones, the bleeding edge buyers are buying for the 9900X or above or wait for the X3D... those buying a 65W low end part likely isn't tinkering much, just like buying 99% of i3 buyers don't bother setting anything

 

[stuff and nonesense]

For me anyone buying the bleeding edge isn't the target for the 9700X and 9600X... they are the latest gen, or a refresh of AM5 cpus, and the lower tier ones, the bleeding edge buyers are buying for the 9900X or above or wait for the X3D... those buying a 65W low end part likely isn't tinkering much, just like buying 99% of i3 buyers don't bother setting anything

Purely bleeding edge because of its newness in this form, it will settle quickly. I agree though.

 

[TheHerald]

There's no contradiction, here. If you were comparing Intel Gen-on-Gen MSRP, then I wouldn't. There is a direct correspondence between each of those products and it's informative to look at what the manufacturer does with pricing of their product tiers.

You're trying to manufacture a controversy where there really is none.


As you're no doubt aware, AMD doesn't regard the 9700X as a direct replacement for the 7700. We'll have to wait and see why not, but maybe they have something planned for the 9700.


Okay, so 22 W vs. 18 W is worth understanding. We'll have to see if this discrepancy holds with newer BIOS. In any case, it's definitely an improvement vs. the 24 W used by the 7700X.


Yes, some reviewers are comparing to the 7700, on that basis. I think that's an interesting comparison, but that the 7700X is the more logical point of reference. Even in that case, the 9700X offers a 12.5% improvement in single-threaded cinebench scores.

You know what I don't get? The 7700 is a better product than the 7700x, since it's cheaper and a lot more efficient, and it comes with an included cooler. So why the heck would you ever compare the 9700x against the worse of the 2 products? The only reason I can think of is if you are trying to portray it better than it is. If that is your goal, sure go ahead, but in that case why is that your goal?

 

[bit_user]

I wonder why. Maybe it has something to do with them having lower power limits. The graph isn't really comparing efficiency, it compares settings.

Caught you in a lie, right there.

It says clearly on that page:

All power measurements on this page are based on a physical measurement of the voltage, current and power flowing through the 8-pin EPS CPU power connector(s), which makes them "CPU only," not "full system." We're not using the software sensors inside the processor, as these can be quite inaccurate and will vary between manufacturers.
...
In this section, we divide the performance achieved by the power usage, to get a Cinebench points per watt single-threaded and multithreaded result.

Source: https://www.techpowerup.com/review/amd-ryzen-7-9700x/23.html
​

So, they are using actual measurements made at the electrical level, not CPU self-reporting!

 

[TheHerald]

Caught you in a lie, right there.

It says clearly on that page:

​

All power measurements on this page are based on a physical measurement of the voltage, current and power flowing through the 8-pin EPS CPU power connector(s), which makes them "CPU only," not "full system." We're not using the software sensors inside the processor, as these can be quite inaccurate and will vary between manufacturers.​

...​

In this section, we divide the performance achieved by the power usage, to get a Cinebench points per watt single-threaded and multithreaded result.​

​

Source: https://www.techpowerup.com/review/amd-ryzen-7-9700x/23.html​

​

So, they are using actual measurements made at the electrical level, not CPU self-reporting!

And the lie is...?

 

[bit_user]

it looks like, since zen 5 doesn't scale much with power,

9700X at PBO Max delivers 15.1% more performance on Blender. I'm not saying it's great, but it's also not nothing.

By comparison, the i9-14900K delivers 24.9% speedup on Blender, between 125W and 253W.

 

[stuff and nonesense]

And the lie is...?

What do you want? Out of the box or tinkered with?

Out of the box gives the possibility of consistency, tinkered with gives no chance. When you benchmark an “optimised” system you are looking for improvement against the stock system.

Settings

CPU test, use the fastest GPU available, get rid of the bottleneck, memory recommended is 6000MTs leave the CPU alone. Leave the bios alone barring xmp/expo.
For (silicon quality notwithstanding) a simple OC, use PBO, the results will be broadly comparable.

Any other variations invalidate the comparison.

 

[YSCCC]

Purely bleeding edge because of its newness in this form, it will settle quickly. I agree though.

I think coz it's the only peek into the Zen 5 lineup, much like if intel release the i3 first and the rest two weeks later. And in all fairness, I am actually surprised that the MSRP of the 9950X at release is lower at $649 vs $699, given how bad the trouble of RPL is haunting Intel, looks like they are trying to eat up some more market share in the time window

 

[bit_user]

Cross-referencing Phoronix non-gaming results with Hardware Unboxed gaming results...

You said it causes regressions in gaming results. Please link (if video, please specify time).

Hardware Unboxed also showed that the real efficiency gain is about 7% more perf at same power usage which is half of what AMD was claiming...

Single-threaded, multi-threaded, PBO on/off? How'd they measure power consumption - full system, package-only; electrical or self-reporting?

TechPowerUp found a 60.2% efficiency gain in Cinebench MT:

Spoiler

As I mentioned above, that was package-only & measured electrically.

 

[TheHerald]

Single-threaded, multi-threaded, PBO on/off? How'd they measure power consumption - full system, package-only; electrical or self-reporting?

TechPowerUp found a 60.2% efficiency gain in Cinebench MT:

Spoiler

As I mentioned above, that was package-only & measured electrically.

Why isn't there a 60% increase between 7700 and 9700x as well? Any ideas?

 

[TheHerald]

What do you want? Out of the box or tinkered with?

Out of the box gives the possibility of consistency, tinkered with gives no chance. When you benchmark an “optimised” system you are looking for improvement against the stock system.

Settings

CPU test, use the fastest GPU available, get rid of the bottleneck, memory recommended is 6000MTs leave the CPU alone. Leave the bios alone barring xmp/expo.
For (silicon quality notwithstanding) a simple OC, use PBO, the results will be broadly comparable.

Any other variations invalidate the comparison.

When im looking at which CPU to buy, and since I care about efficiency, I look at how they perform at the same power, since im going to run them at the same power. When I do that, I find amd's high end chips (well, mainly the 950xs) lovely, but their midrange offerings are an insult, the 9700x not being an exception. That's of course my opinion, you can have yours.

 

[bit_user]

So why the heck would you ever compare the 9700x against the worse of the 2 products?

The 7700 also has lower single-threaded and gaming performance. Yes, it's a compelling product but it's not the one this is said to be replacing. The 9700X is a suitable replacement for the 7700X, so I see no reason not to treat it as such.

I knew it wouldn't be long before you started your shell games with weird product match-ups and running things in highly artificial settings, like an i9-14900K @ 88W.

I think it's telling that you have to find some way to manufacture a controversy and tear down Ryzen 9000. I think it tells us all we need to know about you.

 

[bit_user]

And the lie is...?

I underlined it for you. Here it is, again.

You: "The graph isn't really comparing efficiency, it compares settings."
​

It's about as pure of an efficiency test as it can be. They are not dividing by the TDP or PPT, but rather the actual power used. They labelled it an efficiency test, because that's what it is.

 

[King_V]

Is it a good overclocker? With PBO enabled it draws as much power as a stock 12700k and gets similar performance. That's a 2021 i7 mind you that can be bought right now for 200$.

That's not accurate.
Looking at the power consumption for the 9600X on the graphs with y-Cruncher, Prime95, etc., depending on which test, the 9600X with PBO uses from 108 to 149 watts.

Going back to the same graphs for the i7-12700K, depending on the test, with DDR4, the Intel chip uses from 128 to 188 watts.

With DDR5, that becomes 140 to 195 watts.

In other words, if we give the 12700K with DDR4, it still uses between 20 and 39 watts more than the 9600X at PBO. Or, if you prefer percentages, 17.4% to 26.2% more power draw.

Give it DDR5, and that becomes 32 to 46 watts more, or 29.6% to 30.9% more power.

That is not "as much as" - those are significant differences. And, given that the 9x00X is outdoing the 14th gen Intel chip at most things, but NOT those that heavily depend on multi-threading, how do you expect anyone to believe that the 9x00X has "similar performance" to the 12th gen i7?

EDIT: belatedly realized that @bit_user beat me to the punch on this one.

 

[TheHerald]

I underlined it for you. Here it is, again.

​

You: "The graph isn't really comparing efficiency, it compares settings."​

​

It's about as pure of an efficiency test as it can be. They are not dividing by the TDP or PPT, but rather the actual power used. They labelled it an efficiency test, because that's what it is.

I didn't suggest they divide by TDP. I'm saying what chip scores higher doesn't depend on it's actual efficiency but the out of the box settings. Like the 7950x being on the mid bottom of the graph when it's the most efficient chip in MT performance is a dead giveaway.

The 7700 also has lower single-threaded and gaming performance. Yes, it's a compelling product but it's not the one this is said to be replacing. The 9700X is a suitable replacement for the 7700X, so I see no reason not to treat it as such.

I knew it wouldn't be long before you started your shell games with weird product match-ups and running things in highly artificial settings, like an i9-14900K @ 88W.

I think it's telling that you have to find some way to manufacture a controversy and tear down Ryzen 9000. I think it tells us all we need to know about you.

I think the fact that you are comparing with the 7700x tells us all we need to know about you 😀

 

[TheHerald]

That's not accurate.
Looking at the power consumption for the 9600X on the graphs with y-Cruncher, Prime95, etc., depending on which test, the 9600X with PBO uses from 108 to 149 watts.

Going back to the same graphs for the i7-12700K, depending on the test, with DDR4, the Intel chip uses from 128 to 188 watts.

With DDR5, that becomes 140 to 195 watts.

In other words, if we give the 12700K with DDR4, it still uses between 20 and 39 watts more than the 9600X at PBO. Or, if you prefer percentages, 17.4% to 26.2% more power draw.

Give it DDR5, and that becomes 32 to 46 watts more, or 29.6% to 30.9% more power.

That is not "as much as" - those are significant differences. And, given that the 9x00X is outdoing the 14th gen Intel chip at most things, but NOT those that heavily depend on multi-threading, how do you expect anyone to believe that the 9x00X has "similar performance" to the 12th gen i7?

EDIT: belatedly realized that @bit_user beat me to the punch on this one.

I was comparing to the 9700x. The 9700x with PBO consumes more power than the 12700k

Spoiler

In MT the 12700k and the 9700x , even with maxed out PBO and 170w perform very similarly.

 

[bit_user]

I didn't suggest they divide by TDP. I'm saying what chip scores higher doesn't depend on it's actual efficiency but the out of the box settings.

Now you're gaslighting us. There is no such thing as "intrinsic efficiency". There's only the efficiency of the complete solution, which includes hardware and settings.

I think the fact that you are comparing with the 7700x tells us all we need to know about you 😀

Yes, that I take AMD's word that the 9700X is intended as a replacement for that model.

 

[bit_user]

I was comparing to the 9700x. The 9700x with PBO consumes more power than the 12700k

Spoiler

In MT the 12700k and the 9700x , even with maxed out PBO and 170w perform very similarly.

See? This is exactly what I was talking about. When it's 3.6% more power than Intel, it's "more". However, the performance discrepancy is 10.4% in AMD's favor and yet you characterize that as "similar".

10% is nothing

You're not being intellectually honest. And yet you pretend not to know why people get upset with you. I'll bet the majority of arguments you get into aren't so much about your position as your tactics and how far over-the-top your claims are. Most of us here probably aren't really AMD fans, but mainly just objecting to your unfair characterizations of them.

 

[SunMaster]

.... not sure if that's good nor not....

for the average user, definitely not.

What I do think would be great for the average, midrange, user is to be able to buy a 350-400watt powersupply, knowing it would be plenty for both CPU and GPU.

We're not there yet, unfortunately.

 

[YSCCC]

See? This is exactly what I was talking about. When it's 3.6% more power than Intel, it's "more". However, the performance discrepancy is 10.4% in AMD's favor and yet you characterize that as "similar".

You're not being intellectually honest. And yet you pretend not to know why people get upset with you.

actually now I am thinking he's a hidden AMD promotion agent... his arguements are persuading ppl to go to AMD with everything he argued....

 

[TheHerald]

Yes, that I take AMD's word that the 9700X is intended as a replacement for that model.

Oh im perfectly sure amd would want us to compare it with the 7700x. The question is why would anyone do that?

Now you're gaslighting us. There is no such thing as "intrinsic efficiency". There's only the efficiency of the complete solution, which includes hardware and settings.

Of course there is. The 7950x is more efficient than the 5950x or the 3700x. Yet on those efficiency graphs it scores lower. Are you actually telling me that the 3700x (i have it btw) is more efficient than the 7950x in MT?

Spoiler

 

[YSCCC]

What I do think would be great for the average, midrange, user is to be able to buy a 350-400watt powersupply, knowing it would be plenty for both CPU and GPU.

We're not there yet, unfortunately.

Actually decades ago that is the norm, I still remember in the Pentium D era a 400W PSU is considered excessive. But tbh, with a few still surviving, it still do the job for brosing the web and document processing, we are just advanced too much on the performance front