OcUK Ryzen 2000 series review thread

CAT-THE-FIFTH · 19 Apr 2018 at 23:48

chrcoluk said:
you can do what I did and just uninstall the patch, there is no known meltdown malware in the wild and several other ways to mitigate risk.

Apparently the InSpectre tool can disable it in Windows.

easyrider · 19 Apr 2018 at 23:49

4K8KW10 said:
Won't be done right - still with the chewing gum and still with the useless graphics part on the die. Resulting in extreme heat and lower frequencies and not better gaming performance.

I'm prepared to wait and find out .

4K8KW10 · 19 Apr 2018 at 23:49

Guru3D missed to bench the Ryzens on the very modern and actual gaming resolution of 480p. They somehow included 720p, 1080p and 1440p lol

CAT-THE-FIFTH said:
Looking at the cache latency improvements,they seem significant:

https://www.tomshardware.com/reviews/amd-ryzen-7-2700x-review,5571-3.html

If you check some of the clock for clock improvements in games,it seems more than the 3% AMD touted,and the rest is probably down to the cache latency improvements I suspect.

And the CPU-Z test gains are impressive - north of 36% performance uplift Ryzen 7 2700X 4.4GHz vs Ryzen 7 1800X.

Drollic · 19 Apr 2018 at 23:52

If you're going off popularity, sites should bench pubg, fortnite, cs, dota2/lol, overwatch then maybe a few more titles of various genre such as TW3, PC2 etc. It's not really an issue though with so many review available.

4.4 will be the 5.1/5.2 of 8700k no doubt, won't happen for a lot of people and the ones it does will be running 1.425-1.5v.

chrcoluk · 19 Apr 2018 at 23:54

CAT-THE-FIFTH said:
Apparently the InSpectre tool can disable it in Windows.

it can but my testing on my laptop didnt show a performance gain.

The 8 sec lag I had loading task manager stayed when I disabled meltdown mitigation (but kept patch installed). It then vanished when I uninstalled the patch.

However I havent tested if just disabling improves my desktop FF15 stutters.

Drollic · 19 Apr 2018 at 23:58

Also, can you oc Ryzen in BIOS on x470 according to core loading or is it all core set to a certain frequency in which case, if you clock all core to 42, you're actually hampering the chip slightly under single core use? If this can't be done in BIOS, can it in Ryzen master?

JediFragger · 19 Apr 2018 at 23:58

~>Dg<~ said:
people dont seem to get that intel give us the best gaming cpu for what we need.no bs.no bias.just truth.

That would be all well and good, but now we know their performance was built on unsound foundations (Spectre and Meltdown). The total fallout is still to be seen.

JediFragger · 19 Apr 2018 at 23:59

CAT-THE-FIFTH said:
Apparently the InSpectre tool can disable it in Windows.

Do you have a link for that please? Did a quick google but not conclusive. Thanks

CAT-THE-FIFTH · 20 Apr 2018 at 00:01

Drollic said:
If you're going off popularity, sites should bench pubg, fortnite, cs, dota2/lol, overwatch then maybe a few more titles of various genre such as TW3, PC2 etc. It's not really an issue though with so many review available.

Yeah,a lot of them have been covered in the reviews in the OP. Just seen this one,which did some Apples to Apples testing:

https://www.youtube.com/watch?v=Mr2B0RJd7Nc

Not seen any PUBG or Fornite testing yet though.

chrcoluk said:
it can but my testing on my laptop didnt show a performance gain.

The 8 sec lag I had loading task manager stayed when I disabled meltdown mitigation (but kept patch installed. It then vanished when I uninstalled the patch.

However I havent tested if just disabling improves my desktop FF15 stutters.

This whole Spectre/Meltdown saga has been all rather an annoying affair.

JediFragger said:
Do you have a link for that please? Did a quick google but not conclusive. Thanks

https://www.grc.com/inspectre.htm

JediFragger · 20 Apr 2018 at 00:01

Thanks @CAT-THE-FIFTH Was 99% it was that but wanted to confirm

CAT-THE-FIFTH · 20 Apr 2018 at 00:07

The OP has been updated with some more video reviews.

JediFragger · 20 Apr 2018 at 00:17

CAT-THE-FIFTH said:
Testing Games

https://www.youtube.com/watch?v=Mr2B0RJd7Nc&

I really love the lower CPU utilisation on Ryzen during this vid, shows just how much more headroom it has for longevity :cool:

gavinh87 · 20 Apr 2018 at 00:26

JediFragger said:
I really love the lower CPU utilisation on Ryzen during this vid, shows just how much more headroom it has for longevity

On the other hand, that cpu you just paid for is sat idle.

haszek · 20 Apr 2018 at 00:39

CAT-THE-FIFTH said:
Yeah,a lot of them have been covered in the reviews in the OP. Just seen this one,which did some Apples to Apples testing:

https://www.youtube.com/watch?v=Mr2B0RJd7Nc

Not seen any PUBG or Fornite testing yet

Science studio guy on YT had both.

CAT-THE-FIFTH · 20 Apr 2018 at 00:43

JediFragger said:
I really love the lower CPU utilisation on Ryzen during this vid, shows just how much more headroom it has for longevity

Well I am more interested by the FO4 results,ie,the run they make around Swan's Pond which is in one of the more built up areas of the map. The Sweclockers test is in Diamond City which is more a hub area,but this is more in a dense part of the map. The same channel tested the Ryzen 7 1800X and it was definitely much slower. OTH,even with the Core i7 8700K at 4.4GHZ and the Ryzen 7 2700X at 4.2GHZ,the AMD CPU is quite close. Since the game does not use more than six threads,the Ryzen 5 2600/2600X should be close. I am going to wait and see when the channel compares the Ryzen 5 2600/2600X with the Core i5 8400,and when some more reviews with the game appears to see if the performance improvements hold true(or not).

haszek said:
Science studio guy on YT had both.

Thanks for that - I will add it to the OP.

Hotwired · 20 Apr 2018 at 00:54

B1gbeard said:
All core overclock seems a bit disappointing as well? Not sure what people’s expectations were.

Compared to what.

Compared to all cores at base frequency like the old days the all core max overclock looks good.

But what you see is people pitting simple all core overclock against AMDs adaptive clock system which itself tries to reach maximum clock speeds but in a more flexible way.

That doesn't look as productive because XFR can reach higher frequencies than an all core OC.

So we get these funny results where all core overclocking can mean nothing or be worse because XFR was already doing a better job.

fowler002 · 20 Apr 2018 at 01:06

@~>Dg<~ might be shock to the system but not everyone like mp only games, are are some who couldn’t give two hoots about them. So all these bf, cod, pubg, fortnite etc are irrelevant to them, some of them do care about tomb raider etc. Maybe not ashes of the benchmark, but that’s not the point.

humbug · 20 Apr 2018 at 02:03

JediFragger said:
I really love the lower CPU utilisation on Ryzen during this vid, shows just how much more headroom it has for longevity

Thats really impressive, other than 10 FPS in Arma III there is nothing between the 8700K and the Ryzen 2700X.

chrcoluk · 20 Apr 2018 at 02:06

humbug do you agree the 2600x is a better buy vs 2700x for gaming? Media has its usual frenzy on the highest priced chips but for value for money the 2600x easily has the 2700x beat I reckon.

Also sorry dude, what is that video doing running a 8700k at 4.4ghz? A ryzen clocked up to its limit vs a 8700k nowhere near its limit

Hotwired · 20 Apr 2018 at 02:09

Fascinated to know whose screwup is being highlighted with anandtechs gaming results.

Has no bearing on the product but a disagreement on testing and therefore results will encourage people to be more vigilant when we find out why.