Jump to content
Search In
  • More options...
Find results that contain...
Find results in...


  • Content Count

  • Joined

Everything posted by Mateyyy

  1. I wouldn't buy either of them. The 3080 should be much cheaper, and the 6900XT is a bad deal from the get-go, since the 6800XT offers 90-95% of its performance at 2/3 the price.
  2. And what price is that, exactly?
  3. I reckon the digital monitoring and control of the RMi would be a lot more useful than the marginal improvement in efficiency of the HX. I'd get the RMi. The Seasonic offerings definitely aren't worth the extra cost. What components will you be powering?
  4. I'd maybe call it decent by B450 standards. The Z490-A Pro and Prime Z490-P are generally sitting at around $150. There are cheaper Z490 boards, though they're also significantly worse, so I wouldn't even bother considering them. I don't get what you mean by higher memory speed, the upgrade path is something to take note of if OP will actually need 12/16 cores down the line, and the real-world performance benefit of the increased IPC is certainly not large enough to warrant over €100 more.
  5. It'll work fine. Not having the fastest memory out there certainly isn't the end of the world, and the performance benefit you'd get out of say a new 3600 CL16 kit most likely won't come close to outweighing the cost of a brand new kit of memory. RAM speed not mattering on Intel CPUs hasn't been true for a few generations now.
  6. A decent Z490 board would be ~€30 more than a decent B550, and the price difference between the two CPUs seems to be a bit more than €30. The "newer, better memory controller" is the same as what's used in Ryzen 3000 CPUs.
  7. Strictly in terms of gaming performance, they'll be neck in neck at 1440p. The 10700K will however pull significantly more power, and you'll potentially be looking at spending a little more on the motherboard, so pricing-wise, I wouldn't say the 10700K is better really. I don't know if I'd consider the lack of PCIe 4.0 a con right now, but it could be an annoyance 1-2 GPU generations from now, when PCIe 4.0 might actually make a difference. I'd pick the 5600X, though if you can't wait for it to get back in stock, then the 10700K would also work perfectly fine.
  8. It's not like the 3070 costs less than half of what the 2080Ti did. A refresh would be something like the GTX 770 for the GTX 680, where it's the exact same GPU equipped with some faster memory. Even something like a 2080 Super would be a lot closer to being an actual refresh (though I still wouldn’t really call it that), as opposed to the 3070 and 2080Ti. It's not. The 3070's based on the GA104 GPU, whereas the 3080 is based on GA102.
  9. The SX8200 Pro is the better drive. Which one I would choose however depends on what price they're going for and how much money I've got to spend. ...so?
  10. lmao 2070/2070S-ish..? Just wait for reviews. Why wouldn't they? Assuming you're on a supporting platform, of course.
  11. I reckon an HD 7970 (later rebadged as the R9 280X in 2013) would offer very decent 1080p60 gaming even today, and that's what, 9 years old at this point? GCN in general aged a lot better than Kepler, which is why if I had to go for a GPU from back then, I'd pick AMD over Nvidia (HD 7970 vs GTX 680, R9 290 vs GTX 780). If you're not too bothered about gaming at lower than 60FPS, then even an HD 7870 (later rebadged as the R9 270X) should do an alright job. With older, non-GCN AMD GPUs or pre-Maxwell Nvidia GPUs, you'll potentially be looking at compatibility issues
  12. It's kind of got the look of a Nitro+, but it isn't a 6800XT. https://www.techpowerup.com/review/sapphire-radeon-rx-6800-xt-nitro-plus/2.html To me it looks like an R9 390 Nitro, or maybe an R9 Fury Nitro, which would be a little closer to back when MSI still had their "Krait" lineup of motherboards. You could ask the seller for a GPU-Z screenshot.
  13. Yeah, for some reason I thought that was HWMonitor, lol. Idle temps don't really tell a whole lot, other than cases where there's a bad cooler mount for instance (where you'd see temperatures over 80C), hence why you should run a full load on the CPU. Even something like Cinebench R20/R23 would work.
  14. What are your temps under full load? Also, HWMonitor tends to be quite inaccurate, especially with Ryzen systems. I'd suggest you use HWiNFO64 instead.
  15. I'd say you can do better for $900ish, at least if you're not limited to just Microcenter. $110 seems pretty excessive for an H470 board, there's not much info on the Barracuda Q5 (other than the fact that it does use QLC NAND, which should ideally be avoided for an OS drive), the H510 is a hotbox and I couldn't find anything on the PSU, though there probably is a reason for Redragon not being known for their PSUs (or... for much of anything really, other than cheap peripherals afaik). You could put together either an AMD or a comparable Intel rig for roughly the same price.
  16. How'd spending $40 on a mediocre, unnecessary cooler be more worthwhile than putting that money towards a better PSU/CPU/motherboard? The stock cooler's perfectly adequate for a locked i5.
  17. Or perhaps it's just the effect of RAM caching. OP, do you happen to have "RAPID mode" enabled in the Samsung Magician software?
  18. Those domes sound like they're very satisfying to type on .
  19. OP's already got DDR4 memory in their current system, so there's no point in buying a new kit really.
  20. You can plug just a 4-pin in the 8-pin EPS connector on the motherboard. But, the fact that your PSU doesn't have an 8-pin EPS connector to start with, and that it can only output 15A on the 12V rail (so 180W), is a little alarming, to say the least. Basically, I'd recommend you buy a non-dumpster-fire power supply, instead of potentially creating more problems. You don't need to break the bank to get something decent - there's units like the Corsair CX/CXM, Cooler Master MWE Bronze v2/White v2, be quiet! System Power 9/Pure Power 10 or 11 400W+ (not the 300/350W variants
  21. You know it's proper news when it's just a Wccftech article that starts with "As per our sources".
  22. It outperforms the 1080Ti by a good margin, let alone the 1080 non-Ti. And that's strictly looking at gaming performance. If we're to also take into account feature sets, then the gap between Pascal and newer GPU architectures grows even larger. 300 quid for a used 1080 doesn't sound like a good deal in the slightest if you ask me. At that price, I'd just get a new 5600XT or 2060. If you can't wait a little longer for new GPUs around the price you're after, and you don't mind going used, then at least I'd look for something like a used 5700XT or 2060S/2070, though these
  23. That performance 'difference' is very much within margin of error, and Cinebench barely even cares about memory anyway.