AMD To Handle Boost Clock Frequency Differentials with Firmware Update

Well... wait again to see what's going on with the next update.

 

From the Guru3D article:

"Does this affect your performance?

No, all measurements to date are valid, if there is something going on, then the perf is the same as previously. If a chip is slower, then it also was that during any testing. It works vice versa, results could only become faster. Ryzen 3000 is a complicated processor. AMD already has explained in-depth that there are many variables in play that determine the Turbo single thread bin. The right thread also needs to be prioritized towards the fastest / best core, as not all cores are equal."

And check the 3600-3600X performance difference, which is literally zero. And there is a 200 MHz boost clock difference.

Anyway, if GTX 970 owners didn't care about the 3,5 GB only full-bandwidth memory, people will neither care about this.

 

But they will cry because they are not getting the ghz needed like everyone is just looking at the pc screen to just see what it gets, if its not that its the Vcore problem thats well is fixed but people dont care to seem to look hard to put just a tad bit of work in, I own the 1700 and talking about growing pains with ram,i skipped 2nd gen and went to 3rd for the 3600 noticed i was idle at 1.444 and would not move, i waited and tinker with settings, till i found out that all i needed to do was enable AMD Cool n Quite and vcore drops at idle,

 

AMD is sweating..... trying to avoid that lawsuit.

 

What? 970 had a class action lawsuit against it because people were so outraged.. lol

Or people just want to keep companies honest so we don't get Mobile SoC scenarios where they have boost states for review applications and remove throttling on review revision phones.

 

Well if the other thread is an indicator, AMD fans didn't seem to much care or think anything was a big deal.

 

Well, it's good that AMD is listening to their community. Zen 2 is an important product and they need to do the right thing.

Eh? There were a number of AMD fans who wanted to hold AMD to account over this (me, Denial, etc.).

 

At least AMD intends fix it quickly. Hopefully the BIOS fix really does resolve the boost issues. Mistakes happen but it's what the company does to deal with them that is the real test.

 

There was lawsuit because nVidia originally stated 64 ROPs while cards had only 56.
Analogy would be if AMD cut off one additional core from one chiplet making 15C/30T CPU while selling it as 16C/32T.

Reduced bandwidth while problem in case VRAM utilization went close to 4GB, is not something anyone should have been stressed about as much... since nVidia had strong compression even back then and actual bandwidth does not reach theoretical one for GPUs.

 

Isn't that kind of a bummer that this is the standard though nowadays? It's like everything is early access. Ship it and fix it later.

Their last fix helped introduce the clock speed issue. Will this fix be the one to get the processors to where they should have been on launch day, or will it be months from now? At a certain point you begin to wonder if the company that built the chip knows how it works. You can list a ton of companies that have similar launch issues, but it's kind of crappy that it's expected nowadays.

 

When I said the same thing about Intel and their security issues, AMD fans didn't care how fast they fixed it. Funny how that works....

 

Typical fanboy stuff..... Intel or AMD..... Everybody talking about this like this is some big deal. I don't know why nobody wont talk about memory on mobile phones. My phone has 5684MB RAM (CPU Z, HTC 12+), and it says on GSM arena, on box, everywhere, that it has 6GB???? Every mobile phone i had have, didn't have memory they (company that produce them) stated it has. So, i never heard that someone talked about this. Neither here, neither on sites that are phones specialized. So, my conclusion is - AMD fanboy doesn't care about this "problem", intel fanboy doesn't care about spectre or meltdown, Nvidia fanboy doesn't care about 3,5GB RAM in 970....

 

Looks like that. But If I was to pick smaller memory bandwidth or fewer ROPs. I would rather have smaller memory bandwidth and keep ROPs.

 

Intel does not fix their security holes quickly. Otherwise their benchmarks would not state that not all publicly available patches were applied on their test systems.

At time patch is publicly available, their internal test systems should have them in place already. Especially if they are used for marketing purposes to show end user performance. No excuse there.

Edit: Ups. Double Post.

 

Um, yes they do. Once things were public, they had software fixes out damn quick in BIOS updates - its up to mobo companies to get them out then. Microsoft also had fixes in software. Plus Intel has released hardware fixes in new revisions of newer CPUs. To this day, there are some form of fixes available and not one single damn report of something being exploited. Don't spread BS.

 

You are in thread related to claims that AMD did mislead public. And there are mentions of class lawsuits.
As intel shows benchmarks that have better results than actual client systems. And it is due to intel not applying fixes to their own test systems...