News AMD Ryzen 5 3600 XT Benchmarked Against Intel Core i5-10400

[Admin]

If these benchmarks are to be trusted, the 3600XT won't be winning the battle against Intel's cheapest 10th-Gen Core i5.

AMD Ryzen 5 3600 XT Benchmarked Against Intel Core i5-10400 : Read more

 

[sparrow2]

If these benchmarks are to be trusted, the 3600XT won't be winning the battle against Intel's cheapest 10th-Gen Core i5.

AMD Ryzen 5 3600 XT Benchmarked Against Intel Core i5-10400 : Read more

But is this default setup of the 3600XT or PBO ?

 

[Ninjawithagun]

Did you actually read the article? Tom's stated the benchmarks are most likely fake. Therefore, no conclusions can be made on the true performance of the 3600XT. Stop being an Intel fanboy for 5 seconds and take a deep breath. BTW, I own both an AMD PC and an Intel PC, so I am allowed to talk smack 😛

 

[escksu]

I don't know if the results are real or fake (Tomshardware update that its fake). But I don't think the performance is much of a mystery since its just a slightly faster version of 3600x (100MHz higher boost clock).

We know that its the same Zen2 architecture as 3600/3600x. So only difference is the clockspeed. Of course we don't know the PB/PBO settings, But I personally don't think its that much different from 3600x (likely 100-200MHz higher?) So, overall, I guess we are looking at ~5% increase in performance.

 

[cryoburner]

Update 30/06/2020 4:56 am PT: VideoCardz confirmed our suspicions that the chip pictured by TecLab was not a real Ryzen 5 3600XT, but rather a standard 3600, making this a likely case of a fake preview.

If it were real, why would anyone cover the "XT" part of the product name with tape? That makes zero sense and removes any believability that it might be a real 3600 XT in the photo. Covering the serial number and QR code makes sense, but not the one part of the product name that would identify it as a new processor.

In any case, I wouldn't be all that surprised if they were to get access to a 3600 XT, seeing as the processor is set to launch in a matter of days, and there are likely not only chips in reviewers hands at this point, but also making their way to retailers and system manufacturers. It seems a bit pointless leaking performance numbers with only days remaining though.

Logically, I suspect performance is only going to be a little faster than a 3600X, which is in turn only slightly faster than a 3600. There are still some unknowns in terms of what multi-core boost clocks will be like though, as no specifics have been announced about that yet, and it's possible that the chip could potentially do a better job maintaining higher boost clocks in general.

But is this default setup of the 3600XT or PBO ?

Another thing to ask would be if the 10400 were running at stock memory settings or not. Most people buying one of these sub-$200 locked i5s are likely to pair them with a mid-range B or H-series motherboard, but Intel artificially restricts memory speed on those boards to DDR4-2666. Running an i5-10400 with memory at that speed can cut gaming performance by around 5-10% in CPU-limited scenarios compared to DDR4-3200, so test results utilizing faster memory are not going to be entirely representative of what performance will be like on most systems running that processor.