• 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.

*** 48HR SPECIAL: AMD 3700X ONLY £259.99 !! ***

Zartar117 said:
You don't understand the difference between Coffee Lake/Pinnacle Ridge and Matisse's boost behavior. Coffee Lake/Pinnacle Ridge's single core boost clock is practically guaranteed on any or all cores if the cooling solution and power delivery are good enough. And the boost clock is sustained for much longer (indefinitely if the cooling and motherboard power delivery is good enough). Matisse's boost clock is only guaranteed on one core and it is sustained for much shorter times, so short that many times it is never seen at all. I've tested with a 3600 myself and was let down by the results (one core reaching 4.175 GHz and none of the others going over 4.05 GHz on water cooling running cinebench).

It's still a great CPU but it's clear that the new definition of boost clocks was done to maximize yield at the expense of consistent core quality.

By comparison, all the cores on my 1600X can hit 4.1 GHz when boosting with just MCE (MSI's version) enabled.
Click to expand...

When did you test this 3600, and with what agesa and exactly what awful water cooling solution?
Did you for interest sake, test it with the supplied cooling for comparison?
Given that the wonderful company Intel don't seem to supply an cooling at all now, given their complete and utter ******** TWP ratings.
 
~>Dg<~ said:
how did you know those speeds ? after you brought it. point totally missed. you can debate argue all you want. you dont know what you getting until you brought the product. which shows you how its badly advertised sold from a customers buying point of view. amd love it.
Click to expand...

You don't know what you'll get for Intel either though :/ the reason we roughly know now is because we've seen them long enough, not the same.

With AMD you can say that as well now there averages out there. When first released you didn't know what you would get with Intel.

How can you say it's for one brand and not the other though? Complete rubbish otherwise tbh what you've said. Should both give minimum.all core boost, maybe but the fact is both show base clock and maximum single core boost clock.
 
ordered one of these on Saturday, it arrived on Sunday with MB & RAM as well, everything worked on first try, coming from a retired i5-2500k, this thing's a beast!!!
getting like 25% cpu usage and 100% cpu usage in monster hunter world iceborne XD
should be able to stream a lot better now..
 
FoxEye said:
Well said. A lot of us are happy to see AMD on the rise whilst at the same time sick to death of the fanboys.
Click to expand...

Thanks, I mean look at this reply to my post about the differences in how boost clocks work between Coffee Lake/Pinnacle Ridge and Matisse.

jigger said:
TBH I think it’s you that don’t understand. The 3600X is the higher performance part, but how would the 3600 fair against its Intel counterpart.
Click to expand...

This post has nothing with what I'm talking about, just pointing out that the 3600 is a great CPU (which is literally what I said). But because I criticized AMD's misleading re-definition of boost clock, the fanboy must post off topic stuff about how great AMD's products are (which I agree, I'm pretty happy with my 1600X).

Destination said:
When did you test this 3600, and with what agesa and exactly what awful water cooling solution?
Click to expand...

The same "awful water cooling solution" that's currently cooling my 1600X that can boost to 4.1 GHz on any core. The AGESA was either 1003ABB or 1004. But your post misses the point. I'm not disputing the quality of the 3600, which I think is the new 2500k, I'm talking about the new boost clock definition which differs from the old one in ways that are a bit misleading.
 
Orange Nexus said:
He is claiming that with intel you know exactly what you are getting as advertised. So, i asked, what is the clock of the i9 9900K at stock running Handbrake? Obviously, he does not know cos he does not own one. You need to have one on hand to know or based on reviews. Same with matisse. Their all core OCs are not advertised.
Click to expand...

But with Coffee Lake/Pinnacle Ridge we know that all the cores in any of those CPU's are capable of hitting their advertised boost clocks when the temperatures are under control. With Matisse only one core is guaranteed to be capable of hitting the boost clock. dg hasn't explained his point that well but he's right to criticize AMD on their shady boost clock re-definition.

Orange Nexus said:
BTW, my R5 3600 at stock boost to 4.2GHz in two cores.
Click to expand...

Good for you but there are many of us who have not been that fortunate. Mine could only do 4.175 GHz on one core and none of the others would boost above 4.05 GHz at any given time. Whereas all of the cores on my 1600X can boost to 4.1 GHz.
 
Zartar117 said:
Thanks, I mean look at this reply to my post about the differences in how boost clocks work between Coffee Lake/Pinnacle Ridge and Matisse.



This post has nothing with what I'm talking about, just pointing out that the 3600 is a great CPU (which is literally what I said). But because I criticized AMD's misleading re-definition of boost clock, the fanboy must post off topic stuff about how great AMD's products are (which I agree, I'm pretty happy with my 1600X).



The same "awful water cooling solution" that's currently cooling my 1600X that can boost to 4.1 GHz on any core. The AGESA was either 1003ABB or 1004. But your post misses the point. I'm not disputing the quality of the 3600, which I think is the new 2500k, I'm talking about the new boost clock definition which differs from the old one in ways that are a bit misleading.
Click to expand...

LOL. OK
 
Zartar117 said:
Good for you but there are many of us who have not been that fortunate. Mine could only do 4.175 GHz on one core and none of the others would boost above 4.05 GHz at any given time. Whereas all of the cores on my 1600X can boost to 4.1 GHz.
Click to expand...

R5 3600 PBO+150MHz 4x8GB 3800MHz boosting all cores ~4.3xGHz, link. Stock same application loading CPU ~4.2GHz ACB.

As far as I'm concerned every Matisse chip I have had (5x of) has delivered on clocks. There were 1 or 2 AGESA which had pants SMU FW leading to clocks issues, but AMD has nailed it in later FW IMO.

Enjoyed my Intel setups, but no doubt enjoying my Ryzen rigs more so far.
 
gupsterg said:
R5 3600 PBO+150MHz 4x8GB 3800MHz boosting all cores ~4.3xGHz, link. Stock same application loading CPU ~4.2GHz ACB.

As far as I'm concerned every Matisse chip I have had (5x of) has delivered on clocks. There were 1 or 2 AGESA which had pants SMU FW leading to clocks issues, but AMD has nailed it in later FW IMO.

Enjoyed my Intel setups, but no doubt enjoying my Ryzen rigs more so far.
Click to expand...

Noice. I personally think that the first wave of Zen2 processors were bottom of the barrel bins because of the Rome ramp up and TSMC's 7nm process being a bit immature. I'm sure the current bins are much better. But that's why I think AMD changed their definition of boost clock which was kind of sneaky. They should have renamed it peak clock IMO to avoid confusion.
 
Zartar117 said:
Noice. I personally think that the first wave of Zen2 processors were bottom of the barrel bins because of the Rome ramp up and TSMC's 7nm process being a bit immature. I'm sure the current bins are much better. But that's why I think AMD changed their definition of boost clock which was kind of sneaky. They should have renamed it peak clock IMO to avoid confusion.
Click to expand...

Thanks :) . My R5 3600 is purchased at launch from OCUK, you'll see 1922 stamped on IHS, ie year 19 week 22 produced. I have also owned later produced R7 3700X (1943), as well as 3x R9 3900X from 1944.

Out of all 5x Zen2 the R5 responds to PBO the most. When I tweak BCLK it attains highest FCLK vs all other Zen2 I had. Only the R5 and 1 of 3 R9 could do 4x8GB or 2x16GB at 3800MHz with 1:1:1 FCLK/UCLK/MEMCLK. In my experience no way was it bottom of the barrel, for being early silicon and an R5 3600.
 
Speaking of boosting. I am going to return my 3700X and kept the 3600X as I found it boosted much better and sticks to 4.4Ghz nearly all the time on default settings with no PBO etc.
 
Firestar_3x said:
Can't be bothered to read 8 whole posts, amazing.
Click to expand...

Consider yourself amazed then, it's posts like your little dig just there that justify the decision. However, after trawling through all the rubbish in here, I think the question I was originally asked, not by yourself I might add, was "which of the two I prefer". Now, on Page 3 there seemed to be a question about either a 3700x or a 9900k. I think that was what was being referred to in the question to me, ah the time I didn't feel like wading through the fanboy rubbish to actually confirm that. It would be a 3700x without question, to answer @Troezar .

As for you, no wonder you have almost 30,000 posts with useful contributions like the above. I didn't read the 8 more posts btw :D
 
You must log in or register to reply here.
Back
Top Bottom