- Joined
- Nov 1, 1998
Last edited by a moderator:
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
FRONTPAGE Corsair Dominator Titanium 48GB DDR5-7200 Memory Kit Review
- Thread starter Overclockers.com
- Start date
The memory kit is designed for Intel motherboards (has programmed XMP), but as I mentioned in the review, it works fine on AMD too. I'm still testing it, so you can expect various results on the forums.
Here is a stability test at DDR5-8000 CL36-48-48 1.45V on ASUS Crosshair X670E Gene - most settings at auto.
Here is a stability test at DDR5-8000 CL36-48-48 1.45V on ASUS Crosshair X670E Gene - most settings at auto.
Hello, I bought a kit Corsair like this and I can't get more than 6800MT/s stable on my gigabyte Z690 aorus master.
with xmp1 at 7200MT/s or xmp2 at 7400MT/s without changing anything in terms of timings and voltage, I can log into Windows, but when I run a stability test I have a lot of errors.
MB Gigabyte Z690 aorus master
i9 13900kf
Corsair dominator titanium 7200 2X24gb
with xmp1 at 7200MT/s or xmp2 at 7400MT/s without changing anything in terms of timings and voltage, I can log into Windows, but when I run a stability test I have a lot of errors.
MB Gigabyte Z690 aorus master
i9 13900kf
Corsair dominator titanium 7200 2X24gb
Your motherboard wasn't designed for more than 6400. I remember it couldn't work at more than 6200-6400 with the 12900K CPU. Considering how Gigabyte provides support and new BIOS releases, I doubt that much has changed, but a stronger memory controller in your CPU helps to make some more.
I would set 6800 at tight timings like 32-38-38 or 32-39-39 ~1.40V, as it's not much slower in synthetic benchmarks, and in games you won't see any difference.
I would set 6800 at tight timings like 32-38-38 or 32-39-39 ~1.40V, as it's not much slower in synthetic benchmarks, and in games you won't see any difference.
Since you have 24GB modules, it may work as 32-39-39 or 32-40-40. CL32 may also need a bit higher VDD/VDDQ. Try some various settings, and you will find out.
CR doesn't change as much as in older memory series. You can try CR1, but it probably won't work. It usually works up to 6200-6400, sometimes higher.
Focus on tREFI - usually up to 32k is stable, up to 64k still gives pretty good results, and above 124k is random. Some motherboards don't have more than 64k. You can see in the AIDA64 benchmark how it affects bandwidth and latency.
There are some other timings like tRFC, or tRRDS/L that can help some more. tFAW generally can go down to 16. Some more settings are in my reviews (link in signature or just check last couple of reviews on the front page)- every 2x24GB 7000+ memory kit uses the same Hynix M-die IC and overclocking is almost the same.
CR doesn't change as much as in older memory series. You can try CR1, but it probably won't work. It usually works up to 6200-6400, sometimes higher.
Focus on tREFI - usually up to 32k is stable, up to 64k still gives pretty good results, and above 124k is random. Some motherboards don't have more than 64k. You can see in the AIDA64 benchmark how it affects bandwidth and latency.
There are some other timings like tRFC, or tRRDS/L that can help some more. tFAW generally can go down to 16. Some more settings are in my reviews (link in signature or just check last couple of reviews on the front page)- every 2x24GB 7000+ memory kit uses the same Hynix M-die IC and overclocking is almost the same.
Já estive aqui no fórum assistindo tutoriais sobre timings primários e secundários em DDR5. E neste momento já tenho alguns ajustes feitos com a RAM a 6800MT/s. No AIDA64 tenho 59ns de latência. Eu acho que não é tão ruim assim. Vou tentar estabilizar no CL32. Depois posto os resultados aqui. Mas o que eu queria mesmo era chegar a 7200/7400. Pode ser que o gigabyte resolva isso em uma nova versão do BIOS. Estou fazendo um teste de estabilidade com o Memtest86, mas esse teste leva cerca de 3 horas para ser concluído, existe algum outro programa eficaz e mais rápido?
I've been here on the forum watching tutorials on primary and secondary timings in DDR5 and at this point I already have some adjustments made with the RAM at 6800MT/s.
On AIDA64 I have 59ns of latency i guess it's not that bad. I'll try to stabilize on the CL32 then I post the results here.
I'm doing a stability test with Memtest86, but this test takes about 3 hours to complete, is there any other effective and faster program?
But what I really wanted was to get to 7200/7400. It could be that gigabyte solves this in a new BIOS version.
Sorry for my inlês
On AIDA64 I have 59ns of latency i guess it's not that bad. I'll try to stabilize on the CL32 then I post the results here.
I'm doing a stability test with Memtest86, but this test takes about 3 hours to complete, is there any other effective and faster program?
But what I really wanted was to get to 7200/7400. It could be that gigabyte solves this in a new BIOS version.
Sorry for my inlês
It looks good, and you won't make much more. You can still play with tRRD (sg/dg on the screenshot), tRDRD, and tRDWR, but I don't think it will be much better. I expect maybe 2ns lower latency and 2GB/s higher bandwidth. I don't think it's worth pushing more, and right now, I would run some extended and mixed stability tests.
Somehow I missed your last post. People over the web will try to tell you that their option is the best, but there is no rule, and the best is to use 2-3 different applications for the final stability test. As I said, for reviews, I'm using AIDA64 - memory+cache tests (so it loads the memory controller and other related things), and about 3h is enough.
Recently, I was using HCI memtest for additional tests, but as I said, it's good to run it through a 3rd third-party application that sets additional options. I run ~100% coverage for quick tests and ~1000% for extended.
I wasn't using Prime95, but you set it for ~95% of the total RAM and run for 2-3h.
Memtest86 is quite long, but the last releases are pretty good and run on all CPU threads.
When I test it for my 24/7 PC, I usually run 2-3h of memory stability test, then something like 3DMark in the loop for 1-2h, and then just use the PC and correct something when I notice any instability. Once the PC passes any memory stability test, then is a very low chance of a "bluescreen" or the OS crash.
Typically, the application closes or you get some unexpected application error. Like a specific game may just randomly close. I had that even after passing all stability tests, so I usually run the tests, and when they pass, I set one memory ratio lower, and it always runs without issues 24/7.
Recently, I was using HCI memtest for additional tests, but as I said, it's good to run it through a 3rd third-party application that sets additional options. I run ~100% coverage for quick tests and ~1000% for extended.
I wasn't using Prime95, but you set it for ~95% of the total RAM and run for 2-3h.
Memtest86 is quite long, but the last releases are pretty good and run on all CPU threads.
When I test it for my 24/7 PC, I usually run 2-3h of memory stability test, then something like 3DMark in the loop for 1-2h, and then just use the PC and correct something when I notice any instability. Once the PC passes any memory stability test, then is a very low chance of a "bluescreen" or the OS crash.
Typically, the application closes or you get some unexpected application error. Like a specific game may just randomly close. I had that even after passing all stability tests, so I usually run the tests, and when they pass, I set one memory ratio lower, and it always runs without issues 24/7.
Similar threads
