With the launch of AMD Ryzen 7 5800X3D, we saw how much bandwidth matters in 12th generation Intel processors (DDR4 and DDR5) and today we tested the ZADAK MOAB RGB DDR4 3600C18 2x8GB, which have been sampled thanks to the brand.
We will see its impact on a 12th generation Intel system (Intel Core i9-12900KF) and little by little we will rebuild our DDR4 RAM database, which will be with us for a few more years.
Table of Contents
ZADAK MOAB RGB DDR4 3600C18 – Technical Specifications
The ZADAK MOAB RGB DDR4 3600C18 uses Samsung C-Die chips, a chip that we have not come across so far. Hopefully we will have more documentation on this chip, although it seems to behave quite similar to Samsung B-Die.
It uses a Single Rank configuration and the primary frequency is 3600 MT/s with primary timings 18-22-22-38 via XMP profile. According to the information from the SPD, the PCB of these memories are eight layers and use a voltage of 1.35v to activate the pre-programmed overclock profile.
In addition, being a "gamer product", it has ARGB lighting.
ZADAK MOAB RGB DDR4 3600C18 – Unboxing and photos
Benchmarks and tests in games (720p)
The tests we do in games use a resolution of 720p, so that the bottleneck is not the video card (but the processor) and its relationship with the RAM memories. If you see better results with certain RAM memories (be it positive or negative) it means that it scales with the game.
For the (manual) overclock profile we use 3900MT/s CL16 with 1.5v to DRAM and they went through stability tests, in addition to modifying some secondary / tertiary timings in Gear Mode 1 (to avoid latency penalties).
Before going to tests 720p with games, We will see what results we get AIDA64 Benchmark y Geekbench 4 (for information and reference) and review the testing system we use.
Processor: Intel Core i9-12900KF (Power Limits Unlimited, non-MCE) (https://amzn.to/3Jr6zFC)
Motherboard: TUF GAMING Z690-PLUS WIFI D4 (BIOS 1304) (https://amzn.to/3KzwnAK)
RAM: ZADAK MOAB DDR4 3600C18 2x8GB (Samsung C-Die) <– ASUS XMP-2 (https://amzn.to/3jbm5KX) <—- What we are testing
Graphics card: NVIDIA GeForce RTX 3080 Ti Founders Edition
SSD: Team Group Delta MAX 250GB
SSD #2: Silicon Power A55 2TB (https://amzn.to/3NT4vtq)
Closed Loop Liquid Cooling: Lian Li Galahad 360 (https://amzn.to/3uvzOCO)
Power supply: EVGA SUPERNOVA 750W P2 (https://amzn.to/371dCaP)
Operating system: Windows 10 Home 21H2 (Win Game Mode On, RSB On, HAGS OFF). We'd rather test with Windows 10 for now.
Overclocked (manually) 3900 "MHz" CL16
Overclocked (manually) 3900 "MHz" CL16
Overclocked, the MOAB RGB DDR4 3600C18 by ZADAK show an improvement in 14.67% a comparison of its factory performance, in the synthetic test Geekbench 4 (memory section).
720p gaming tests
Why test games in 720p? It is the first question that many will ask. The answer is quite simple. If we test at a higher resolution, the bottleneck is directed entirely at the video card (in this case, the NVIDIA GeForce RTX 3080 Ti).
We want to see how it scales games to a lower resolution and what a difference the impact of RAM frequencies makes. During the review of the AMD Ryzen 7 5800X3D (processor that does not require super memories to work quite well) we tested Alder Lake-S (Core i9-12900K/KF) using DDR4.
Both DDR4 and DDR5 Intel processors benefit from increased bandwidth and reduced latencies by having a smaller amount of L3 Cache.
720p will show us the improvements between one RAM or another, when the bottleneck is the temporary memory that links several devices of a PC, also known as RAM memory.
A statistic reminder...
Before detailing our configuration system, let's refresh a little what they are AVG FPS, 1% LOW y 0.1% LOW.
AVG FPS (Average FPS): As the name says, it is the average number of frames per second within a specific sequence. It is the most used measure, but it does not detail the whole story, since there are FPS drops.
1% LOW: Within an entire frames per second dataset, the 1% LOW is the value equal to the lowest 1% within the frames dataset (of the specific sequence ordered ascending). In simpler terms, it is the frame where you see the FPS drop that exists within a specific sequence.
-Assassin's Creed Origins
-FFXV (Not GameWorks)
-Middle-Earth: Shadow of War
-Shadow of the Tomb Raider
-The Witcher 3
There are certain games where RAM scales significantly in various titles with overclocked RAM. The improvement in Assassin's Creed Origins en Average FPS is a mere 7%, by contrast, the 1% LOW is almost 15% higher, getting more stable frametimes and minors.
There are other games that are going to get quite similar results between different RAM kits, because the bottleneck is the video card (yes, even at 720p Ultra) or because they have an in-game maximum limiter. An example of limitation by play is Control.
Another title like Death Stranding, the improvement is minimal with the ZADAK MOAB RGB DDR4 3600C18 because it is also limited to 240 FPS.
Games like Final Fantasy XV, Shadow of War, Shadow of the Tomb Raider y The Witcher 3, is better appreciated because it is worth overclocking RAMs manually. The sample you gave us BACK it scales pretty well (and we didn't spend too much time overclocking and validating it).
Getting the most out of a RAM memory game can take several days and even weeks, to validate stability and get every % of performance. The improvement in these previously mentioned titles is approximately 14% more FPS using the same processor.
Once again, don't forget that as the game is more restricted to a video card bottleneck, you will see a smaller scale of performance improvement relative to RAM versus processor in use.
Conclusion – Slightly better performance than 3200 MHz CL14 on Alder Lake-S, overclock very well
The results of the memories ZADAK MOAB RGB DDR4 3600MHz CL18 2x8GB are slightly better than a setup of 4 DIMMs 3200 MHz CL14 (G.Skill FlareX). This result is quite impressive and differs a bit from what was previously tested (at least with Alder Lake-S). Apart from latency, games benefit from the bandwidth offered by RAM with Alder Lake-S processors.
This earns our seal of approval with 12th generation processors.
El result that impresses us is in overclocking manually (I made a quick one that took 1-2 days to validate) to 3900 “MHz” CL16 with 1.5v. That means that there is some room for improvement in secondary/tertiary timings to squeeze more performance out of the kit we have today.
The Samsung C-Die chips work quite similar to the B-Die apparently (I have no information on their difference) but the performance improvement in games that they benefit from is quite noticeable with manual overclocking.
Be careful, DDR5 memories are the ones that get the most out of 12th generation Intel processors, but they are substantially more.
It is the first product that ZADAK has offered us and has really impressed uso.
We have another kit of them to test (which should work in Gear Mode 2) so we will see the difference in performance with the other higher clocked RAM.