13900KS review

[Deleted member 258511]

but you said really overclockers buy intel... you need to make your mind up old boy

Yes, if you like tweaking and overclocking Intel is the way. In that persons use case he was wondering should he wait to see what the 3d CPU’s are like before jumping the gun. No overclocking preference was mentioned.

Have I offended you in some way?

 

[humbug]

But everything you said applies to intel as well, you think xmp tunes the timings?

Again, I've said it around hundreds of times, why are we talking about instead of just...testing it ourselves? I mean I know why, cause nobody with a 7950x dares to run the games and post a video, but honestly that tells me all I need to know about the legitimacy of pcgh's review. I've personally tested a 7950x with tuned 6200 ram and his cyberpunk numbers are completely spot on in cyberpunk lows for example - there is an actual 40% difference between the 7950x and a 13900k. It is what it is, and it if wasn't, plenty of people with the 7950x around could post their results and prove otherwise.

I mean check this, with DLSS performance the run is completely cpu bound and the framerate is absolutely horrible

Cyberpunk 2077 4K | RTX 4090 | Ryzen 9 7950X | Psycho Settings | Ray Tracing | DLSS ON & OFF

Buy Gamekeys at CDKeys ►https://cdkeys.pxf.io/xkJg3v Please hit the thumbs up button and subscribe if you like what you see.Recorded with my dedicated RECORD...

www.youtube.com

Well yes, that is what XMP does, don't tell me it doesn't... part of qualifying a kit for the QV list is to test and confirm all the timings from the manufactures settings, if it isn't and you set those XMP speeds the timings go in to a (sloppy) safe mode if timings haven't been qualified for set speed in the BIOS. Why do you think AMD send RAM and motherboards with new CPU launches and ask "please use these for your review"? because they are a combination that have been qualified and it matters.... A lot.

Its a discrepancy of results we are talking about, and reasons for it, you can intentionally or unintentionally manipulate results by quite a chunk just by playing silly vagina with the ram. its very easy to do.

I've said over and over again the results are legitimate, but its no more legitimate than HUB results, can we agree on that?

 

[Bencher]

I've said over and over again the results are legitimate, but its no more legitimate than HUB results, can we agree on that?

Absolutely. But you think hub gets different results because of ram. But that is not the reason. The reason is he is running lighter settings. Cyberpunk without RT is incredibly lightweight for CPUs, and hub runs with it off. Even on spiderman he is using RT at the medium settings (high and scale to 6) but even with those settings, the difference is massive.

Think about this for a second. PCGH was running a 12900k with 4400mhz DDR5, and it was basically matching all zen 4 across the board. Doesn't that tell you anything? I've tested my 12900k maxed out at 5.4 ghz with 6000c30 tuned ram and it was still around 12 to 15% slower than my STOCK 13900k with 7600c34 ram. I assume we agree that an overclocked to the wall 12900k is faster than stock zen 4, and yet it was far behind a stock 13900k. That tells you all you need to know about the difference between zen 4 and RPL.

Check my video, it's my 12900k running Spiderman with hub's settings. In his benchmark he scored an average of 122. Now look at what im getting, and add around 15% on top of that for a stock 13900k.

12900k Spiderman Hwunboxed settings (High RT 6)

12900kSpiderman framerateRT30901080 dlssSpiderman ddr5

www.youtube.com

 

[humbug]

Absolutely. But you think hub gets different results because of ram. But that is not the reason. The reason is he is running lighter settings. Cyberpunk without RT is incredibly lightweight for CPUs, and hub runs with it off. Even on spiderman he is using RT at the medium settings (high and scale to 6) but even with those settings, the difference is massive.

Think about this for a second. PCGH was running a 12900k with 4400mhz DDR5, and it was basically matching all zen 4 across the board. Doesn't that tell you anything? I've tested my 12900k maxed out at 5.4 ghz with 6000c30 tuned ram and it was still around 12 to 15% slower than my STOCK 13900k with 7600c34 ram. I assume we agree that an overclocked to the wall 12900k is faster than stock zen 4, and yet it was far behind a stock 13900k. That tells you all you need to know about the difference between zen 4 and RPL.

Check my video, it's my 12900k running Spiderman with hub's settings. In his benchmark he scored an average of 122. Now look at what im getting, and add around 15% on top of that for a stock 13900k.

12900k Spiderman Hwunboxed settings (High RT 6)

12900kSpiderman framerateRT30901080 dlssSpiderman ddr5

www.youtube.com

And that's where we come right back around to complete the circle, like a snake eating its own tail the only consensus that you will agree with is Intel good / AMD bad, so the only reason for the discrepancy is that everyone else is doing it wrong, and they uniquely are right, because only with that reasoning can you come back to this the fundamental need that you have.

Its the same way you dealt with the power efficiency question, its the same way you deal with everything Intel, they have to be better.
It reminds of websites like Userbenchmark and Principle Technologies, its full of mad gymnastics to get or argue desired results, they know its mad, we know its mad, they know we know its mad, they know we think they are insane over a CPU brand, but they persist anyway, because they are.

 

[Bencher]

And that's where we come right back around to complete the circle, like a snake eating its own tail the only consensus that you will agree with is Intel good / AMD bad, so the only reason for the discrepancy is that everyone else is doing it wrong, and they uniquely are right, because only with that reasoning can you come back to this the fundamental need that you have.

Its the same way you dealt with the power efficiency question, its the same way you deal with everything Intel, they have to be better.
It reminds of websites like Userbenchmark and Principle Technologies, its full of mad gymnastics to get or argue desired results, they know its mad, we know its mad, they know we know its mad, they know we think they are insane over a CPU brand, but they persist anyway, because they are.

Well, once again, easy way to settle it, ive asked around 9999 times now. The fact that no one is willing to test tells me all I need to know, and it should tell you as well if you weren't a tiny teenie bit biased

 

[humbug]

Well, once again, easy way to settle it, ive asked around 9999 times now. The fact that no one is willing to test tells me all I need to know, and it should tell you as well if you weren't a tiny teenie bit biased

Most people are smarter than me and don't involve themselves with something that is obviously circular attrition.

 

[Bencher]

Most people are smarter than me and don't involve themselves with something that is obviously circular attrition.

Of course, posting a run to prove or disprove the review is terrible, I get it

 

[humbug]

I'll let you in on something.

If Intel and AMD was a man you meet in a pub, you would say both are C word and T word, but AMD more T word because through experience and sheer will Intel have C word nailed down far more than AMD who come across more T word because they awkwardly think they need to be C word but can't pull it off making them so much more clown vagina of the equation.

 

[Grim5]

What

 

[IT Troll]

C = CPU, T = Tile?

 

[LtMatt]

I'll let you in on something.

If Intel and AMD was a man you meet in a pub, you would say both are C word and T word, but AMD more T word because through experience and sheer will Intel have C word nailed down far more than AMD who come across more T word because they awkwardly think they need to be C word but can't pull it off making them so much more clown vagina of the equation.

 

[Deleted member 258511]

I'll let you in on something.

If Intel and AMD was a man you meet in a pub, you would say both are C word and T word, but AMD more T word because through experience and sheer will Intel have C word nailed down far more than AMD who come across more T word because they awkwardly think they need to be C word but can't pull it off making them so much more clown vagina of the equation.

What the???

 

[Frenzy]

I think he's trying to say he's picking the lesser of two evils, I think they're as bad as each other personally.

 

[Deleted member 258511]

I think he's trying to say he's picking the lesser of two evils, I think they're as bad as each other personally.

Keep checking back hoping to see an actual review from a customer on here, not that I expect many people to buy these of course, but this thread has gone in a total different direction.

 

[Robert896r1]

Keep checking back hoping to see an actual review from a customer on here, not that I expect many people to buy these of course, but this thread has gone in a total different direction.

What do you want to know? Our group has a bunch of them already. They're just binned K sku's. IMC is still a lottery (since intel doesn't bin for IMC). If you have a golden K, no need for a KS.

Here's some binning stats one of the teams did:

 

[Deleted member 258511]

What do you want to know? Our group has a bunch of them already. They're just binned K sku's. IMC is still a lottery (since intel doesn't bin for IMC). If you have a golden K, no need for a KS.

Here's some binning stats one of the teams did:

That’s more than enough information. Pretty much what I expected to be honest.

The price difference between the K and KS isn’t too big though, so if I was in the market for a k I’d probably just get the KS.

Did your group find similar results with the 12900k and KS?

 

[Frenzy]

That’s more than enough information. Pretty much what I expected to be honest.

The price difference between the K and KS isn’t too big though, so if I was in the market for a k I’d probably just get the KS.

Did your group find similar results with the 12900k and KS?

I was thinking the same. To be honest I would have bought by now because I'm kind of tired of comparing stuff, problem is that there's no stock on certain bits - stock estimated mid Feb, this means I may aswell wait for the x3d results. Not expecting anything wild compared to their last x3d parts and I'm sure they would have been leaking results if it was anything special.

There's a good 13900k on mm of you aren't fussed buying second hand.

 

[RSR]

Did your group find similar results with the 12900k and KS?

You can't compare SP numbers for the 12900K and 12900KS as it has different VIDS so the SP scores are different.

The 13900K/KS you can compare the VIDs, so the SP's are a lot closer.

I'll be putting my 12900KS SP (P) 106 up for sale soon as well, that's a very good sample.

 

[RSR]

What do you want to know? Our group has a bunch of them already. They're just binned K sku's. IMC is still a lottery (since intel doesn't bin for IMC). If you have a golden K, no need for a KS.

Here's some binning stats one of the teams did:

Ive seen quiet a few chaps with SP (P) 128 and 129's now

Also the IMC does seem to be better from the samples I have seen on the KS CPUs as well.

 

[Frenzy]

Ive seen quiet a few chaps with SP (P) 128 and 129's now

On the K? What site? I want to look at the data that's got to be a very rare sp...

On other overclock sites they seem to be saying the k and ks results can't be compared, so hmmm