• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

13900KS review

Poneros said:
AMD defence force big mad right now. Saddest part is they don't even get anything out of it, this is a volunteer service! :cry:

And all because they can't get to grips with objective, verifiable empirical data.

laugh-laughing.gif
Click to expand...
Well it's fine to be questioning the review, it's the part where they don't actually post their own numbers from their own computer that gets me puzzled. Whenever I call out a review for being crap (I have done so in the past with tpup) i post benchmarks from my computer to show the issue. They don't, they just disappear. Which frankly makes me realize they know the numbers are fine, they just don't like them.
 
humbug said:
You know, they keep banging on about "AMD Fanboyism" Intel are the only ones with people who have set up entire Intel Fanboi websites masquerading as legitimate reviewers, bear that in mind the next time you get called that for disagreeing with the narrative by people who keep posting links to them.
Click to expand...
There are some very pro AMD YouTube channels also to be fair. I’m no Intel cheerleader either, but to say Intel are the only ones doing such things isn’t right.
 
The RAM speed talking point has been debunked a million times. What's funnier is that the same people who spout it are also the same people who cry rivers about using 720p to test, but the only way you can ever even see a difference thanks to memory speed increases is at 720p...
As for what Hardware Unboxed is saying, they're straight up incompetent as far as CPU testing goes and also fall under the category of people described above. I'd be shocked if they actually had a test for a game where memory speed made such a huge difference, particularly for Zen 4. Can't honestly think of any, especially as they test at 1080p and with canned benchmarks and at suboptimal settings.

dxwXJIZ.png

edit:
 
Last edited:
Robert896r1 said:
I'm just laughing at the 100mhz "OC" over stock. I wonder how much unnessary voltage they pumped into that.
Click to expand...

yeah lol

How many us were buying binned 9900k 5.3 chips from (the now defunct) silicon lottery that ate power like no tomorrow eventhough the 5.2 chips were available? :D

The present situation is not sustainable but Intel wanted the extra revenue from binning their chips to a greater extent rather than, giving it to silicon lottery (and others) and well, here we are. The anandtech article running 7950x at 65w versus the 13900k was an eye opener.

Where does that leave the 13900ks? I would sooner it existed than not, at least we get an intel factory warranty but it was never intended to be an efficiency king at stock.

Undervolting it and capping the wattage at 125w is where my interest would lie with this chip.
 
Besty said:
The anandtech article running 7950x at 65w versus the 13900k was an eye opener.
Click to expand...
Eye opener indeed, but not for the reasons you think. The 7950x at 65w was consuming over 90w while the 13900k at 65w was actually consuming 71w. The numbers im quoting are taken from their review.

I don't know, people think there is a huge gap in efficiency, but there isn't. In heavy absolute MT performance like cbr23 and the likes yes, the 7950x has an advantage at equal wattage, but it's not big. 10-15% tops. But that doesn't extend down the stack, in lower models intel has the absolute advantage in efficiency in the midrange models. But that's only looking at heavy MT workloads. In mixed workloads RPL poops all over zen 4 in efficiency even with unlimited power limits. Im talking about actual productivity workloads, not nonsense synthetics. Autocads, premiere, photoshop and the likes, rpl just walks all over zen 4 in both performance and efficiency. And the gap is HUGE, the 13900k is 50%+ more efficient than the 7950x at those workloads. Don't take it from me

www.youtube.com

INTEL & AMD both lied! REAL WORLD power consumption is MESSED UP

Intel & AMD have both 'lied' about their real world power consumption. Here's the Truth:Sponsored Message:Whokeys 25% discount code : TN20 Windows 10 Pro O...
www.youtube.com www.youtube.com

This guy was a huge proponent of zen 3 for creative workloads, many people were calling him an amd shill, but after alderlake he started suggesting intel instead cause they are so much better
 
RavenXXX2 said:
Click to expand...

I get why HUB used 6000 vs 6400, because AMD told them 6000 was the sweet spot, so fair enough.

But if you're running under 6000Mhz then you should be running them apples to apples, you shouldn't be running one slower than the other, if the Ryzen system was running 6000Mhz with tight timings and the Intel system running 5600Mhz with loose timings there would be an outrage, they would never do it.

Yet they feel comfortable enough doing it the other way round.

This is why their results differ so much from everyone else...
 
Last edited:
humbug said:
I get why HUB used 6000 vs 6400, because AMD told them 6000 was the sweet spot, so fair enough.

But if you're running under 6000Mhz then you should be running them apples to apples, you shouldn't be running one slower than the other, if the Ryzen system was running 6000Mhz with tight timings and the Intel system running 5600Mhz with loose timings there would be an outrage, they would never do it.

Yet they feel comfortable enough doing it the other way round.

This is why their results differ so much from everyone else...
Click to expand...
Intel can run 7600+ ram, amd can run 6200 if you are lucky. Why would you use the same ram? You are handicapping intel that way.
 
Bencher said:
Intel can run 7600+ ram, amd can run 6200 if you are lucky. Why would you use the same ram? You are handicapping intel that way.
Click to expand...
That would have been better.....

HUB used 6400 on the Intel system and 6000 on the AMD system, i don't have a problem with that, i already said this.

What Computer Base De did is not an equivalence to that.
--------------

I knew something was off given their results are very different to everyone else, when you have an outlier like that no rational thinking person tries to steelman it and elevate it above all others by postulating they do it better than everyone else, at least now we know what that something is, and i'm not saying its illegitimate or fake, but it is a worst case scenario for AMD as its set up to exaggerate its weaknesses.
That in its self should make any rational thinking person at least question their motives.
 
Last edited:
I'll say this, Intel running out of money is in my view a good thing, i can't wait for them to bleed even more because they haven't bled enough, YET, i think everyone would benefit from that.

Keep doing what you're doping Intel, keep throwing that money around.
 
Last edited:
humbug said:
I'll say this, Intel running out of money is in my view a good thing, i can't wait for them to bleed even more because they haven't bleed enough, YET, i think everyone would benefit from that.

Keep doing what you're doping Intel, keep throwing that money around.
Click to expand...
For anyone wondering why, this is a known trait of Intel, and they don't do it to provide you with a monopoly of cheap CPU's and meaningful innovation year on year. Think back just a few years.
 
Last edited:
humbug said:
That would have been better.....

HUB used 6400 on the Intel system and 6000 on the AMD system, i don't have a problem with that, i already said this.

What Computer Base De did is not an equivalence to that.
--------------

I knew something was off given their results are very different to everyone else, when you have an outlier like that no rational thinking person tries to steelman it and elevate it above all others by postulating they do it better than everyone else, at least now we know what that something is, and i'm not saying its illegitimate or fake, but it is a worst case scenario for AMD as its set up to exaggerate its weaknesses.
That in its self should make any rational thinking person at least question their motives.
Click to expand...
You are basically saying Intel is paying them to test each CPU at its stock out of the box settings. And that makes sense to you?
 
Bencher said:
You are basically saying Intel is paying them to test each CPU at its stock out of the box settings. And that makes sense to you?
Click to expand...

It depends on the RAM timings and if the RAM they used is on the board QV list, because if it isn't the timings are basically slop even if you apply XMP.

I know that much.

I bough some Vulcan Z 3200Mhz, not on my boards QV list, i knew by default they would be slow as hell, but i did some research, they have the same Micron-E IC's as the next cheapest kit at £130 at the time, which was on the QV list, mine were £90, Corsair LPX, even the 3600Mhz variant had the same IC's but with loser timings, 18-20-22 vs 16-18-18.

Sure enough when i plugged them in it gave me 18-20-20 timings with sub timings that were basically junk, for example trfc was at 920, with some tweaking i got all the timings to the same level as the LPX kit at 3200 and 3600, i'm actually running them 3800 18-21-21 with sub-timings about twice as tight as the XMP at 3200, that trfc for example is now at 540, from 920 at 3200.

The difference in performance that's made is between 20 to 30%.

I don't know what the timings they used, they don't say, but the performance is way down and given that Ryzen responds the same way to RAM speed as it does just tight timings and their results are so far off everyone else i'm going to agree with what Steve Walton suspects, its been ran with junk timings, the Mhz is far less relevant.

That doesn't make it illegitimate, but it does make it a whole lot of contrived crap.
 
Last edited:
humbug said:
It depends on the RAM timings and if the RAM they used is on the board QV list, because if it isn't the timings are basically slop even if you apply XMP.

I know that much.

I bough some Vulcan Z 3200Mhz, not on my boards QV list, i knew by default they would be slow as hell, but i did some research, they have the same Micron-E IC's as the next cheapest kit at £130 at the time, which was on the QV list, mine were £90, Corsair LPX, even the 3600Mhz variant had the same IC's but with loser timings, 18-20-22 vs 16-18-18.

Sure enough when i plugged them in it gave me 18-20-20 timings with sub timings that were basically junk, for example trfc was at 920, with some tweaking i got all the timings to the same level as the LPX kit at 3200 and 3600, i'm actually running them 3800 18-21-21 with sub-timings about twice as tight as the XMP at 3200, that trfc for example is now at 540, from 920 at 3200.

The difference in performance that's made is between 20 to 30%.

I don't know what the timings they used, they don't say, but the performance is way down and given that Ryzen responds the same way to RAM speed as it does just tight timings and their results are so far off everyone else i'm going to agree with what Steve Walton suspects, its been ran with junk timings, the Mhz is far less relevant.

That doesn't make it illegitimate, but it does make it a whole lot of contrived crap.
Click to expand...
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

www.youtube.com

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 www.youtube.com
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom