AMD to Revise Specs of Ryzen 7 9700X to Increase TDP to 120W, to Beat 7800X3D

[R0H1T]

well, the public deserves the nonsense companies pull, I suppose.

Well the public is stupid, they have an attention span of Lenny G & they like Marvel comedies tragedies for some reason

 

[AusWolf]

As for increasing the TDP, for the last time TDP is not power draw

It's not, but it can relate to it. Power limit = TDP x 1.35 on an AMD CPU.

Sure, there's nothing that says the 9700X will actually reach its power limit (just like the 7800X3D doesn't), but if it's anything like the 7700X, then with good enough cooling, it will.

The other thing is that TDP is an arbitrary requirement for cooling on AMD. An uprated TDP usually means you need better cooling, even if your actual power consumption isn't higher.

genuinely asking, where is 9800X?

Probably the same place where the 7800X is. Nowhere.

 

[R0H1T]

The actual (max)power draw can be lower, what increased TDP means is that the chip gets to clock higher for longer if there's enough thermal headroom & an appropriate workload. The power draw still depends largely on the chip you have & of course the rest of the system.

The other thing is that TDP is an arbitrary requirement for cooling on AMD.

Yes & I think they changed it at least once since the initial Zen release? Not sure about that but the complex formula is stupid in general.

 

[AusWolf]

The actual (max)power draw can be lower, what increased TDP means is that the chip gets to clock higher for longer if there's enough thermal headroom & an appropriate workload. The power draw still depends largely on the chip you have & of course the rest of the system.


Yes & I think they changed it at least once since the initial Zen release? Not sure about that but the complex formula is stupid in general.

That's right. But then, that gives people reason for concern, imo. Of course TDP isn't power draw. But it is related to power limit, which may be related to power draw, which all are somewhat related to cooling.

 

[R0H1T]

Well AMD aren't the sharpest tools going around, they're going for efficiency crown in mobile/servers & literally the opposite for desktops? This race to be the next P4 or Dozer is dumb & old

 

[Onasi]

@R0H1T @AusWolf
Essentially, the formulae and how AMD defines their TDP is, really, practically irrelevant in this case. The fact that they, apparently, decided to nearly double the stated TDP for the 8-core model just means that they decided, for some bizzare reason, that potentially increasing the available power and, presumably, thermal envelope for the chip is “worth it” if they can present better numbers in Day One reviews. This, to me, just shows a lack of confidence in the product. I think it’s a poor look, but again - the public cares more than it should for minimal benchmark differences, so I guess everyone is dumb in this case.

 

[Carillon]

If true, the x3d parts will be closer to the regular zen5. Or they will come out with high TDPs x3ds.

 

[Klemc]

A new power plant has to be build.

 

[Wasteland]

If true, the x3d parts will be closer to the regular zen5. Or they will come out with high TDPs x3ds.

It's an interesting point. A move like this might suggest that AMD has decided against releasing a Zen 5 x3d chip at all. Or that they're worried about how long it would take them to release one.

 

[Onasi]

If true, the x3d parts will be closer to the regular zen5. Or they will come out with high TDPs x3ds.

Not necessarily. X3D chips are traditionally weird. I would assume that the trend of “faster in games, slightly slower in everything else, very low power despite the equal or higher TDP” will continue with Zen 5 too. They can’t really make “high TDP” X3Ds, not significantly high at least, to my understanding. The cache is sensitive to temps.

It's an interesting point. A move like this might suggest that AMD has decided against releasing a Zen 5 x3d chip at all.

Naaaaah. It’s just petty vanity bullshit and, perhaps, some inside knowledge on how Arrow Lake will perform.

 

[AusWolf]

@R0H1T @AusWolf
Essentially, the formulae and how AMD defines their TDP is, really, practically irrelevant in this case. The fact that they, apparently, decided to nearly double the stated TDP for the 8-core model just means that they decided, for some bizzare reason, that potentially increasing the available power and, presumably, thermal envelope for the chip is “worth it” if they can present better numbers in Day One reviews. This, to me, just shows a lack of confidence in the product. I think it’s a poor look, but again - the public cares more than it should for minimal benchmark differences, so I guess everyone is dumb in this case.

Not necessarily. It may retain similar power characteristics, but with a higher limit for extreme cases, or just with a bigger cooler requirement badge (because TDP on AMD is just that).

Let's not forget that the 7800X3D eats between 80-90 W under full load, but has a 120 W "TDP" that comes with a 162 W power limit.

Edit: If it's truly a 65 W part, then a bundled Wraith Stealth cooler should be able to run it. Perhaps AMD realised that this is nowhere near reality.

 

[Klemc]

Not necessarily. It may retain similar power characteristics, but with a higher limit for extreme cases, or just with a bigger cooler requirement badge (because TDP on AMD is just that).

Let's not forget that the 7800X3D eats between 80-90 W under full load, but has a 120 W "TDP" that comes with a 162 W power limit.

Not sure of my though, but if an x3d is coupled with a RTX 5000>6000, it will be pushed to his max more often right ?.. with appropriate games.

 

[Onasi]

@AusWolf
I already mentioned that X3D are weird. So comparing to them is pointless. 7700X, on the other hand, had a PPT of 142W and, lo and behold, that’s more or less exactly what it was hitting at full load.

Not sure of my though, but if an x3d is coupled with a RTX 5000>6000, it will be pushed to his max more often right ?.. with appropriate games.

Nah, the 7800X3D remains a 50 watt part even in CS at 1080p low with a 4090. If that doesn’t push it - nothing will.

 

[Dragokar]

I don't think they're going to change the 9700X, but if they are so eager, they can push in an 9800(X) or 9700XT without changing the 9700X.

 

[AusWolf]

Not sure of my though, but if an x3d is coupled with a RTX 5000>6000, it will be pushed to his max more often right ?.. with appropriate games.

That's only if the game can fully utilise 16 threads, which I highly doubt. But even then, you're looking at 80-ish W max.

@AusWolf
I already mentioned that X3D are weird.

On the contrary. Power-wise, it works just like an old quad-core i7. What's weird is AMD's TDP and PPT and the 1.35x multiplier between them.

 

[Onasi]

@AusWolf
In that case you ARE correct, but I was talking it being weird by AMDs own standard. But yeah, their whole TDP/PPT scheme is unusual and I don’t actually remember if they ever explained how and why they came up with the 1.35x multi for that. They just did, put that fact out there and operated on it ever since.

 

[P4-630]

So they basically did an intel move...

If it works it works right?...

 

[EatingDirt]

Ah, the good old “crank the power up to win in benchmarks” move. I would have thought AMD would be smarter than this, but apparently not and they’ve resorted to cribbing from Intels playbook. A mistake, IMO, but seeing how a lot of people reacted in the thread about regular Zen 5 not beating X3D Zen 4 chips in gaming like that was a warcrime worthy of a Hague trial… well, the public deserves the nonsense companies pull, I suppose. Hopefully, they would leave in the old PPT settings as a pre-set option a la Eco mode.

Is it a mistake? They want to actually sell the CPU. No one is going to consider it if they can just get a 7800X3D that performs better. As you said, if someone is concerned about power, ECO mode exists as an easy solution for those concerned about power draw.

I really don't understand why people care. These CPUs are unlocked, you can configure them however you want. That's like caring about the out of the box brightness of your TV. Whatever

Indeed. With AMD's ECO mode, you can easily configure the CPU to a lower TDP. It's a sacrifice. You can have lower power draw, or you can have marginally better performance. I see them boosting the available power draw, and subsequently performance, it as a win-win.

genuinely asking, where is 9800X?

I'm assuming AMD has decided to abandon the non-X3D *800 CPU line. It makes sense from tier standpoint, leaving no confusion on which CPU's are the better CPU's for the average-joe consumer.

 

[Dimitriman]

Ah, the good old “crank the power up to win in benchmarks” move. I would have thought AMD would be smarter than this, but apparently not and they’ve resorted to cribbing from Intels playbook. A mistake, IMO, but seeing how a lot of people reacted in the thread about regular Zen 5 not beating X3D Zen 4 chips in gaming like that was a warcrime worthy of a Hague trial… well, the public deserves the nonsense companies pull, I suppose. Hopefully, they would leave in the old PPT settings as a pre-set option a la Eco mode.

I would look at the positive side of this, a lot of people also complained that it didn't beat the 7800X3D and having it at 65W always seemed too string in my opinion.

 

[Onasi]

Is it a mistake? They want to actually sell the CPU. No one is going to consider it if they can just get a 7800X3D that performs better. As you said, if someone is concerned about power, ECO mode exists as an easy solution for those concerned about power draw.

I don’t know, is it? From my perspective - kinda. Nobody sane upgrades every generation anyway. And those who are interested in gaming performance ONLY are also those who will wait for the X3D models. The real draw will be the productivity and general performance increase and in those categories the 7800X3D would have lost even without the TDP increase seeing as how it’s slower than even the 7700X.

People seem to forget that X3Ds are narrow-focus tools. They do gaming exceptionally well. That’s it. For everything else they are inferior to their same-gen and DEFINITELY next-gen counterparts. Before someone flips the lid, “inferior” doesn’t mean BAD in this case, but it is what it is.

I would look at the positive side of this, a lot of people also complained that it didn't beat the 7800X3D and having it at 65W always seemed too string in my opinion.

That’s what I assume, yeah, hence the last sentence in my post you’re quoting.

 

[Chomiq]

When not simply release 9800X.

 

[JustBenching]

@AusWolf
In that case you ARE correct, but I was talking it being weird by AMDs own standard. But yeah, their whole TDP/PPT scheme is unusual and I don’t actually remember if they ever explained how and why they came up with the 1.35x multi for that. They just did, put that fact out there and operated on it ever since.

Their explanation is that some of the heat is dissipated to the sides of the cpu so the cooler doesn't need to dissipate the whole PPT but just the tdp.

 

[Rais]

Ah, the good old “crank the power up to win in benchmarks” move. I would have thought AMD would be smarter than this, but apparently not and they’ve resorted to cribbing from Intels playbook. A mistake, IMO, but seeing how a lot of people reacted in the thread about regular Zen 5 not beating X3D Zen 4 chips in gaming like that was a warcrime worthy of a Hague trial… well, the public deserves the nonsense companies pull, I suppose. Hopefully, they would leave in the old PPT settings as a pre-set option a la Eco mode.

I get your point, but in desktop application even 50w of power poses no issues from delivery to thermal management, so what is the point of a 65w 9700? You can still deliver it as a power level, and still do a good impression on bench. In the ends, sales are important.

 

[napata]

Their explanation is that some of the heat is dissipated to the sides of the cpu so the cooler doesn't need to dissipate the whole PPT but just the tdp.

But that doesn't make sense because the cooling you need depends on a ton of factors. If TDP actually had anything to do with cooling then you'd find different formulas for different core counts.

Say you use AMD's TDP of 105W for both a 7950x & 7700x. The cooling you need is vastly different for each CPU to not thermally throttle at that setting. The 7950x can be cooled with a mediocre air cooler, maybe even a low end one. The 7700x needs a top end cooler to not thermally throttle. If you follow AMD's interpretation then they'd need the same cooling which is just nonsense.

 

[R0H1T]

You can still deliver it as a power level, and still do a good impression on bench. In the ends, sales are important.

You know there's a reason this image is fan favorite of Anime/manga nerds & PCMR

Spoiler