VonZigmas

Is the GRAM temperature monitored by any sensor so I could check if they are overheating? It didn't sound like the card struggled to keep the temps down or anything.

Basically I had Blender open, along with some YouTube and other pages in the background. All of a sudden the screen goes black and all I could notice were some brown lines in the corner. This image I found sort-of describes it, except they were possibly horizontal and definitely not covering the whole display. The thing is, it happened so fast, you could barely notice it. I'm sure it didn't last a second. Nothing crashed, YouTube didn't even seem to cut off the audio. Everything came back as sudden as it went dark. Blender seemed to be more zoomed in than before, but maybe I was in the middle of doing that anyway. By now I've looked through the components for any visual damage, particularly capacitors, but everything I could see seemed fine. Although looking at the 390 PCB now, there are quite a bit more of them than the single line you can see through the heat sink. Still, am I right in thinking that even a single bad capacitor would be noticeable via more constant symptoms? So far putting on some load with Far Cry 3 shows no issues. Took out my PSU for a closer look as a cap seemed suspect, but it appeared to be the same yellow-green dab of paint that the all the others are marked with, just a bit thicker in the crease. Also checked if everything's seated nicely too. Anything else worth checking? Things to look for in case it happens again? What are the chances this was a software glitch and not a hardware issue? Any similar, known issues with this combo or any separate piece of software? Just really not liking the idea of having hardware failure in this 'still kinda new, but old enough to be out of warranty' stage of ownership. The probably relevant stuff: Sapphire R9 390 NITRO / Driver version - 18.2.2 i5 4460 ASRock H97M PRO4 Seasonic M12II-620 EVO Rest is in my profile.

They're not the cheapest things around as far as the main stuff goes (Xbox, PS, Steam), but you can get considerably less expensive ones like the Logitech linked above and they might work just as well for you. Wireless is also not necessary, which saves you from getting batteries and an adapter. I'd probably just get an Xbox One controller alone and then decide if the USB cable really annoys me enough to warrant spending money on a wireless setup. Also boy do keyboards get more expensive than that..

Same with a PS4 controller. Although I do recommend a wireless setup of some sort. I was gonna go the USB route if I went for an Xbox controller, but since the Steam Controller came with its own receiver, I really can't imagine having a wire running from it at all times.

Well that's the problem of profiles, not getting XInput non-Steam games to work. Besides, that hardly seems a problem. Steam offers some of their own templates, which I found work just fine for, say, Minecraft with no tweaking. And on the topic of Minecraft, I just noticed there ARE community profiles for it, which I imagine to be the case for a lot of other popular titles too.

Is that not solved by adding the games to your steam library and launching them that way?

Having almost zero previous experience with controllers in general, I got a Steam Controller on the last winter sale and I've been liking it a lot. Lots of customization options and works great as just a mouse for Windows. Supposedly better for FPS games, which is why it might work better for GTAV as well (and maybe Dark Souls too?). I've only tried GTAIV with it for a very short while, but didn't like the controller experience there in general. Could see myself getting better in Portal 2 though, so maybe just some getting used to needed. Mostly use it for racing games at the moment. Back paddles are great to have and the dual-stage triggers may be very useful too. An Xbox or a PS4 controller can apparently be set up to work basically the same way via Steam, so it's more down to which one you're used to more (or what features you like better).

Yeah, it's not really necessary to note it (but still good to know and not any less accessible than anything else in the tech specs IMO), the thing here is that they DO have a "The LLC resonant topology with DC to DC ensures complete compatibility with all modern day PCs." line on all their non-M CX models.

You'd think they'd mention it somewhere on the page in that case. In fact, all grey CXMs omitting that feature from their description, along with a "2015 Edition" slapped in the title make for an odd marketing choice in comparison to the non-M units.. My bad though.

Huh? http://www.corsair.com/en-us/cx-series-cx550m-550-watt-80-plus-bronze-certified-modular-atx-psu-na Besides existing, it's also group regulated, far as I could find, which would put it no higher than the same T4 here and in some places costs more than the non-modular 2017 CX550.

Yeah, the CX450 has only a single PCIe connector listed. But yet again, I can't find whether that's 6 or 6+2 pin. Still, shouldn't they be separate? The 770 requires a 6 pin and an 8 pin to power it. Assuming Corsair means 6+2 connectors by 'PCIe', that should work? Why do you think so?

Looking for a PSU for a GTX770 (Gigabyte Windforce 3X OC) and an i5-4570. Arrived at two options: Corsair CX550 (CP-9020121) Cooler Master MWE 550 (MPX-5501-ACAAB) CX550 seems to be T3, where does the MWE stack up? Both appear to be DC-DC and are basically identical in price at ~58€. The only clear difference from the specs is that CM includes OCP, whereas Corsair doesn't. Does that alone make Cooler Master the better option? I can't find reviews for the latter (or former? but since it's tiered, I guess I'm missing something?), any ideas? Possibly other options for the price? Also, so they've got two PCIe connections then? Is it me or does this make no sense? Either way, CX550 is listed as having 2 PCIe connectors on their site (written as 2x 6+2 pin PCIe in a local store) and CM has 2X PCI-e 6+2 Pin on their page. Just to confirm, that's fine for the card then?

Guess it was easier to market. And capacitors are replaceable, I don't suppose there's a chip with an identical number of pins that has the missing features?.. Gotcha. So why would he say OCP is meaningless in most cases as if acting like +5V and +3V don't exist? Seems quite a large number of older single-rail PSUs omitted the feature. Then there's SCP, which should catch a short circuit and I'm not finding a particularly clear answer as to what else could cause a massive overcurrent without tripping any other protections (are there no fuses, for example?). Extra protection if SCP fails, I get that, and from what I've gathered - something about shorts in ICs not always being detected by SCP? Yet is there ever zero load on a rail? My understanding is that both tests already performed the absolute edge cases of load/no load and the unit doesn't go out of spec, even if it does come close. Just a quick look at Sapphire's page. Regardless, 397W under load doesn't lessen my point. Assuming this is correct, CPU and GPU could be drawing around 520W alone. I'd really have to rely on that above peak performance with a 450W supply. 550W should work, yeah, but even then it doesn't give too much to spare. I was already going for the "less power, better quality" deal with the M12II (in comparison with some 700+ PSUs) and I believe there were worse routes to take. I don't know where the G-550 was at the time, chances are it just wasn't available. Also likely that, since they were both in the same tier according to some, I went with the fully modular unit while also giving myself the extra 70W for peace of mind. What I'm mostly getting from this is that tiers suck and I should do research myself. If it only it were actually this black and white..

So why would they not include UVP on +12V? Is there a semi-logical reasoning, like +12V being not as sensitive to undervoltage? Or was it probably just a cheaper IC? And how bad is it to not have OCP at all? Doesn't OPP work instead? Quote from http://www.tomshardware.com/reviews/power-supplies-101,4193-21.html - "In single +12V rail PSUs, where OCP is meaningless in most cases, OPP takes over its role and shuts down the PSU in case the +12V rail is overloaded." So for both reviews the biggest deviation happens on +12V at near 5%, which requires for there to be maximum load on +3.3V and +5.5V and minimum on +12V. According to gecid's review, "However, for the sake of fairness, it should be noted that such a value is achievable only on a laboratory bench, since this requires the maximum load of the channels + 3.3V and + 5V, and the voltage + 12V, on the contrary, should be used at minimum. Obviously, in real operating conditions the situation will be absolutely opposite.". Would a couple 2.5" drives really max out the +3.3V and +5.5V rails? And under what conditions would there be minimum load on +12V while that's happening? Even in such a case, 5% equates to.. 12.6V? (is that what it translates to?) Which doesn't seem like too big of a deal. Since it's all within specs, I'd assume hardware is designed to handle it? I don't think so actually. I'm running an R9 390 NITRO. Sapphire themselves suggest a 750W supply. I've been told the 620W Seasonic is enough to handle it (and it handles it just fine in my experience), which is one of the reasons why I got it. A 450W PSU sounds a bit iffy when the GPU alone is rated at 375W (at max load I presume). Can't say I push it to the limit all that often, or maybe the GPU and all the other fans just drown out the PSU fan, to where it doesn't stand out to me. If there are two similarly reviewed products, I'll feel better going with the manufacturer that is generally known for good products. That's my simple thought process. Warranty and post-purchase support matter too, as things go wrong even on the absolute best devices. No, one shouldn't go for the lesser thing because of the brand name, and I feel like I've read enough to know that Seasonic aren't flawless (which isn't surprising, as probably no brand is), but I just think reputation matters, more or less.

The 620W one. So while it is group regulated, the impression I get from all I've read is that, as far group regulated goes, it's still a pretty decent unit. Another thing, is that there's really no proper review for the EVO version of it. The consensus seems that it's the same unit, just made fully modular. But can we confirm that it's really identical and performs the same? I believe the semi-modular one didn't have enough PCI-E 8 connectors for my card, while the EVO version did. Something must've been changed? And what about the protection IC? As I can't find much good or wrong about it. What protections does the unit lack? Under what scenarios it being group regulated could harm me? What's realistically the worst that could happen? Noise-wise, I found it to be just fine. Not really any louder than anything else in my system (which, to be fair, isn't that quiet and maybe why the PSU doesn't bother me). Only trying to run a Metro benchmark at 4K really made it sing. I think it actually does matter, to an extent. No, you should never blindly buy something just because the manufacturer has a good reputation, however good reviews upon that 'good reputation' do make a difference in my opinion. Most who got Seasonic units seem to be happy with their purchase. To quote STRMfrmXMN's whitelist - Seasonic: "Literally everything. Everything. You are buying "Safe"sonic, even on their lower-tier stuff.". I think that may be going a bit far, but he clearly knows more than I do. I bought it, because that's what I happened to choose on what's still my first build. Not planning to go exclusively with Seasonic in the future if there are clearly better options available, but so far I had a good experience, albeit very limited. Was just wondering why thought so and mostly what exactly I'm in for myself.