DDR-RAM Hardware Reviews System

Successful first attempt? – Acer Predator Talos DDR4-3600 CL18 2x 8 GB Memory Kit | Test with Benchmarks and Teardown

Let’s finally take a look at the performance, first in the synthetic benchmarks AIDA64 Cache & Memory version 6.32.5600 and Geekbench 3 version 3.4.3. All benchmarks were run three times and then averaged from the results.

The read and write speeds scale almost linearly with the clock rate, which is why there are no big surprises here. Even in Gear 2 with DDR4-3800, the Intel system can still set itself apart from the XMP configuration here. Accordingly, the completely manual setting on the AMD platform cannot yet show its advantage in the timings. The control pattern, like its clock speed, also sorts itself out behind the DDR4-4400 configuration.   

Copy and latency tests are known to be relatively closely intertwined, as low latencies allow faster switching between writes and reads, and thus more copying. On the Intel platform, the configuration with DDR4-3800 and Gear 2 is almost 5 ns behind its peers on the same platform. This nicely shows the uncanny valley of latency on Rocket Lake between DDR4-3733 and DDR-4000 – you’d rather stay either below and in Gear 1 or above in the Gear 2. 

The B-Die kit can score points in latency with its naturally tight timings, but loses out to the DDR4-4400 configuration significantly in copy due to the clock speed and loose remaining timings. On the AMD platform, the improvements from one setting to the next are relatively small, but steady. With the manual DDR4-3800 1:1 setting, even the Gear 2 latency of the 11700K can be caught up with at the same clock speed. 

Geekbench 3 is known to be a good all-round benchmark and relatively reliable indicator for performance in real-world applications. Lower latencies and higher clock rates contribute almost equally to the result here and thus carry the DDR4-4400 setting almost to the 10000 point mark. Rocket Lake is truly a rocket booster for RAM performance. On the AMD platform, the Talos kit with manual timings can even outperform the B-Die control kit.

Kommentar

Lade neue Kommentare

Gurdi

Urgestein

1,370 Kommentare 900 Likes

Wow die Differenzen in Cyberpunk sind schon heftig zwischen den Profilen.

Antwort 1 Like

RX480

Urgestein

1,841 Kommentare 840 Likes

Könnt Ihr bitte mal noch den Stromverbrauch der CPU in CP2077 messen?

Wäre mal interessant ob Gear2 trotz OCen sparsamer/gleichhungrig beim RKL ist.(vgl. mit Gear 1)

Antwort Gefällt mir

Gurdi

Urgestein

1,370 Kommentare 900 Likes

Also CP brauch ordentlich CPU Leistung, mein 5900er zieht da perma 100 Watt.

Antwort Gefällt mir

RX480

Urgestein

1,841 Kommentare 840 Likes

Jo,
jetzt werden schon die AiO`s mit Push+Pull ausgeliefert:
(könnte bei mir noch 2x140er außen draufsetzen als Pull... momentan aber net erf. mit dem 9400F@50W)

Antwort Gefällt mir

skullbringer

Veteran

306 Kommentare 328 Likes

hab das mal interessehalber mit HWinfo rausgemessen, Resultat fällt aber aus wie ich es mir schon gedacht hatte: maximal 2 W Unterschied zwischen dem selben RAM Setting in Gear 1 vs. Gear 2

Also letztendlich innerhalb der Messtoleranz, zumal ich auch nicht genauer mit Hardware auf den 12 V Schienen messen kann.

Antwort 2 Likes

RedF

Urgestein

4,646 Kommentare 2,545 Likes

Ohne Plastik sehen sie wirklich besser aus. Mit wirken sie einfach nur billig.

Antwort Gefällt mir

Danke für die Spende



Du fandest, der Beitrag war interessant und möchtest uns unterstützen? Klasse!

Hier erfährst Du, wie: Hier spenden.

Hier kannst Du per PayPal spenden.

About the author

Xaver Amberger (skullbringer)

Werbung

Werbung