Yeah ASRock also allowed for some real fun numbers.
The heat coming off it at that level would be insane. I’d hold off until some independent reviewers can run them through their paces. All we know at the moment is based on leaks and AMD marketing material. Neither method is known for fully informing the consumer. There’s bound to be a gotcha lurking in the bushes.
4 Likes
Of course I’m not going to try it: I don’t know the first thing about C++. You’re being extremely antagonistic just because I was ripped off by a company you like.
@Zavar While I agree with you, Threadripper Pro and EPYC systems don’t make sense in a nonmoney-making situation, in a money-making endeavor, you have to do a cost-to-benefit analysis and a return on investment calculation to figure out which would be the better investment. Invest in upgrading 4 or 5 PC to a 7950X or replace the current systems with one Threadripper Pro or EPYC server. You also got to remember to use distributed. Rendering adds complexity to whatever workload you are doing.
I wouldn’t be surprised if AMD decided to also Change the chipset (AM6) at either the end of 2025 or the beginning of 2026. So instead of having five years on AM5, we will have three years on AM5 instead of five years on AM4.
170W is 170W of heat no matter what clock speed you reach.
Don’t confuse socket and chipset. AM4 had a lot of chipsets…
1 Like
I’m not exactly counting on it but I wouldn’t be surprised if we see the arrival of AM6 in 2025 as you say however the same Zen(gen n) Core Complex Die(CCD) that will first arrive for AM6 might also get packaged and released for AM5 perhaps giving effectively a last upgrade path for AM5ers at the ‘cost’ of not getting to fully benefit from the new core arch/clock improvement and obviously sticking with the older gen IO die(maybe half gen upgrade to the latter).
I expect this to be an opportunity AMD might take advantage of if there’s sufficient competition to push them to go for it, but unless the stars align exactly right it probably won’t happen. Makes too little sense.
Except 170W TDP =/= 170W of heat
Consumers don’t understand that TDP is for OEMs and AIBs.
Yes it’s 230W PPT - but that still doesn’t change how power, heat and temperature works.
So 170W TDP somehow converts to 230W now? Physics doesn’t change that you referenced a thermal design profile as power draw, they do not mean the same thing.
I’m trying to help you understand that you have solutions and that there’s nothing “fictitious” about the specs of the chip. But you’ve apparently tried nothing and are out of ideas, so just complain on forums instead I guess.
It can be. AMD doesn’t usually overshoot their TDP in actual power draw margin but they’re often very close.
The 230W figure is the package power limit for AM5. It’s not the TDP of any chip so far announced, but it is the maximum amount of power they can pull from the board. The reality is we don’t know yet how closely AMD aligned power draw to their TDP, but historical evidence indicates that real power draw could be close to the rated TDP.
Because you’re wrong, and all you have is insults.
I’m gonna have to ask you stop being so antagonistic and stop derailing the thread
He has not thrown a single insult and has only provided facts
I’d like to see proof otherwise
4 Likes
Most of the 5800x I saw were hitting 5Ghz out of the box in regular B550 motherboards or x470 motherboards…
1 Like
That is not true at all for at least the last two full generations of Zen, max power draw is almost always more than TDP and just under PPT except in the lower end SKUs. Intel is worse but AMD is still 35% higher than TDP rating on the all of the R9s.
I don’t have a horse in this race and I definitely regret getting involved, but the only point I was trying to make was that if someone is going to quote TDP, at least quote PL2/PPT figures as those are the true max power draw numbers that modern processor do actually hit when allowed to. Especially now that motherboard manufacturers essentially allow for unlimited PL2/PPT boost. My Z490 Unify defaulted to 4096 seconds on the PL2 and when set to a water cooler the max power draw was also set to 4096W. An extreme case maybe, but becoming more common each generation. Anyone planning on getting into high end Zen4 better prepare for 220W sustained loads if you want to get the most performance out of your chip.
My apologies if I offended anyone and for the snarky tone.
1 Like
So you know this, why the earlier reply?
65 TDP is 88W PPT
105TDP is 142W PPT
170TDP will be 230W PPT according to most info.
1 Like
Lets not forget AMD’s TDP = Thermal Design Power. I.e. the heat the chips give off/loss of energy as heat. Of course PPT would be higher.
Well… sort off, not at all. AMD Ryzen TDP Explained: Deep-Dive on TDP Definitions & What Cooler Manufacturers Think | GamersNexus - Gaming PC Builds & Hardware Benchmarks
PPT is what turns to heat if you reach full usage.
1 Like
Oh wait, hold on. So that’s the boost frequency? I saw two sets of frequencies. That’s confusing, then. Why not just show the base frequency? So it’s not actually 4.5GHz?
In any case, AMD have really gotten so much momentum since they launched Ryzen. I wonder what Zen 5 will be like in terms of power draw and IPC.
And sheesh. DDR6? If Nova Lake doesn’t launch with DDR6 i’ll pop a fookin blood vessel. It’s like when I was waiting for DDR5 to come out. Then the Rona happened. 
And as far excess energy turning into heat. 
Could they one day make use of that heat? Could we somehow harness it for something? Like an attachment to the cooler that goes out the back of a case if you’ve got grometted holes for water cooling? it’s basically a long tube that starts all the way from your CPU cooler to the back of the case. Then extends out and is attached to a cup holder. So it can keep your hot beverage hot?