• 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.

AMD Zen 2 (Ryzen 3000) - *** NO COMPETITOR HINTING ***

TrM

TrM

Associate
Joined
3 Jul 2019
Posts
744
if you subscribe to the overhyped gouge ram - that is the b-die...
micron e-die is nearly just as good and less than half the cost, and can easily get 3200c14 speeds overclocked
not that ram speeds are a big performance determinant for ryzen 3000 anyway...

I brought a set of Corsair vengeance rob pro cl I8 22 22 22 42 ram from a competitor which has micron e die only tried at rated speed 3600 but managed to eat it stable at 17 17 17 37 which is better then I thought I would the ram cost less the 100 pounds. I was gonna get the dominator rgb ones but they was 70 pounds more and I couldn’t quite afford it after the 3900x and X570 upgrade
 
Permabanned
Joined
23 Apr 2014
Posts
23,553
Location
Hertfordshire
Associate
Joined
21 Sep 2018
Posts
895
The more extensive tests dont show much improvement at all between 3200 tight and other higher speeds, if any. So for gaming at least, that expensive RAM money would be better spent elsewhere.

Of course this doesnt apply to people with no budget. :p

You got to take note of the timings in these tests. A 3200 CL14 might just be as good as a 3600 CL16 or a 4000 CL18. Im not advocating to buy expensive ram, but for some reviewers to declare that fast ram and tight timings do not affect performance of Ryzen 3000 need a closer scrutiny. That's all im saying.

The game cache and the ipc improvement may be all that we need. We can sacrifice ram.
 
Soldato
Joined
18 May 2010
Posts
22,376
Location
London
interesting test result, albeit just 1 test only.
i do wonder why then, that the 3 reviews i've seen so far tend to show (minimally) increasing fps when getting faster/tighter ram.
warrants a good looking at.

I was reading earlier the benefit to a lower cas latency is when data is being swapped in and out rapidly. Random IO basically.
 
Man of Honour
Joined
26 May 2012
Posts
16,474
Pls link one that used tight timings. Not 3600 CL16.
the two i've seen are with 3733c16. tech report uses looser timings, so lets discount that then.
3733c16 = 8.57ns latency
3533c14 = 7.93ns latency

would dropping the latency by 8% give a 10-12% boost in SOTTR benchmarks?
as i said previously, at this early stage of ryzen 3000's life cycle, i'd be more weary if all reports come back congruent. i'd expect some to show faster/tighter ram = better, some to show otherwise. (basically the whole point of meta-analyses lol)

i'm curious now...
 
Associate
Joined
21 Sep 2018
Posts
895
the two i've seen are with 3733c16. tech report uses looser timings, so lets discount that then.
3733c16 = 8.57ns latency
3533c14 = 7.93ns latency

would dropping the latency by 8% give a 10-12% boost in SOTTR benchmarks?
as i said previously, at this early stage of ryzen 3000's life cycle, i'd be more weary if all reports come back congruent. i'd expect some to show faster/tighter ram = better, some to show otherwise. (basically the whole point of meta-analyses lol)

i'm curious now...

From that it is obvious that 3600 CL16 will have higher latency. A 3466 CL14, which can be easily reached by a 3200 CL14 ram, might even be better than that 3733.

So, we can agree that more tests are needed and going by reviewers analysis of effect of ram cannot be relied on.

I made a bobo. sorry. Earlier we have a member who shared his finding in Cine15. From DOCP 3600 (i assume CL16) to 3600CL14, he gained 40 pts.
 
Man of Honour
Joined
26 May 2012
Posts
16,474
So, we can agree that more tests are needed and going by reviewers analysis of effect of ram cannot be relied on.
we can only go by what is available in front of us at this point. like smoking (cancer), thalidomide (fetal malformations) to name but a few.
as newer evidence comes to light, then we'd know more.

at the moment, we have:
1) one test with all over the shop ram showing minimal improvement
2) two tests that shows an improvement, albeit minimal, going up ram speeds
3) one screenshot on one benchmark showing decreased latency giving a hefty boost to performance.

PS: linky to the source of that pic?
 
Associate
Joined
21 Sep 2018
Posts
895
we can only go by what is available in front of us at this point. like smoking (cancer), thalidomide (fetal malformations) to name but a few.
as newer evidence comes to light, then we'd know more.

at the moment, we have:
1) one test with all over the shop ram showing minimal improvement
2) two tests that shows an improvement, albeit minimal, going up ram speeds
3) one screenshot on one benchmark showing decreased latency giving a hefty boost to performance.

PS: linky to the source of that pic?


Here tam, post 830

https://forums.overclockers.co.uk/threads/zen-2-owners-thread.18858946/page-42

Here the Cine 15 Results, post 22371 (ref to 22366)

https://forums.overclockers.co.uk/t...3000-no-competitor-hinting.18825852/page-1119
 
Back
Top Bottom