TheXDS

But, why isn't AIDA64 doing the same, just for sake of testing?

... Could it probably be a misconfigured SATA controller? I mean, it could be in RAID mode, and Windows might not have the drivers for it. It could be that the SATA ports are disabled in the BIOS settings. There's no such thing as enabling a Hard drive... as far as I know... Let us know how it goes...

(Hey, if Luke is reading, I guess this would actually make an interesting video for The Workshop...) Well, what do you know, AIDA64 and Prime95 tacke the CPU rather differently. Not susprises at all. I was doing some overclocking one night and did run a few stress passes with Prime95 on my CPU. After that, I got to do some testing with AIDA64 (because it allowed me to do some stress testing of several things at once) and notticed how the temperatures were different between both programs. Prime95 hit my CPU considerably harder than AIDA64, like... Hard. I'm running an AMD FX-8120 dinosaur, at 4.25 GHz, idling at 45 C. Prime95 brought it to around 75 C. and, AIDA64, did "only" reach 68 C. As of now, I only tested those 2 programs. do you have any suggestions on other software that I could use to Stress my CPU? I also did a similar experiment to stress test my GTX 970. And, didn't really nottce much difference first. I wanted to generate as much heat as possible, and hit every transistor inside that poor chip. So, I fired up Furmark, alongside GPU-z stress test, as well as Valley, As well as several test at once from MSI afterburner and EVGA, and AIDA64, and even put Blender to render a complex surface with complex shaders. It was literally a mess. My GPU didn't really throttle, but all of the different workloads caused the GPU to output less heat (probably because some the many workloads were lighter, so the heavier ones would take less GPU clocks per second to be performed)

by *new* I meant that it didn't have much data written to it, it was pretty much the OS it came with, the recovery partition and, that's it. And, in the other hand, I've used and abused 40 GB HDDs even today, and they still rocking like new. BTW, the issue with the 840 Evo is well documented. A nasty firmware bug lead to several problems while reading, so the performance would literally drop below 10 MB/s in a matter of weeks. Samsung and some other OEM manufacturers released several patches for it, and the last one did solve the slowdowns, but led the drive to constantly rewrite the data, so the lifespan was shortened for reads as well as writes. I'm not a defendant of the HDD, but at least in my experience, it has been less of a hassle to deal with when they fail. Edit: Write performance was fine, and newly written files would read fast, but not as fast as the drive should (around ~300 MB/s when it should be ~500 MB/s on a 6 month old drive) and I didn't even write that much to it. It only contained my OS, and a few programs I use a lot, but my documents and my Steam library was on another HDD. Again, I want to reiterate, when I bought it, it was almost new, and I didn't know about the issues with the firmware.

Hello, Well, I know this particular thing has been discused a lot on almost every tech channel. But, I really want to know if the tradeoffs given by an SSD versus a regular RAID0 array of disk (ignoring the power draw) would matter enough to get one. Long story short, I bouught an almost *new* SSD from a guy who was upgrading his computer. I'm not sure what computer he had, but I ended up with a Samsung MZ7TE512HMHP-000L2 SSD, which is basically an OEM version of the dreaded Samsung 840 Evo (you know, the one with the horrible slowdowns). As being based on it, my *new* SSD also had those issues. I didn't know about it then. Well, six monthis passed, and I notticed that my computer would slow down considerably, and from time to time the AHCI controller would hang for a minute or so. Of course, I was really worried, and after a whole day clicking on every article and forum I could find about my issue, I realized that my SSD had this problem, and it would be fixable... If only my drive wasn't OEM. There's no firmware tools for my drive, even if it's the exact same thing as an 840 Evo. So, I got in contact with the guy, and he said that he will try to help me out in getting another drive somehow, maybe through warranty or someting. Well, right now I made myself a RAID0 array of HDD I had laying around, and yes, it's pretty loud, and doesn't really give me the seek performance of an SSD, but It got me to thinking... Is it, from a reliability standpoint, really worth having an SSD over an HDD? I mean, both can fail and stop working at all, and yes, HDD's are very fragile, but ther are phisically fragile, not datawise. Or at least not that much. If well taken care of, a Hard disk will outlive the rest of the components of a computer, whereas an SSD will last for a few years before it gets to an unusably slow state, or even completely failing. I'm totally aware that if one of my RAID0 disks dies, my data is gone. But so will if my SSD decided to stop working completely (which was pretty close to do anyways) So, would you thinkk this theme would do for a good TechQuickie video? I would like to know all the pros and cons of an SSD over a closely performing RAID0 array of disks (again, ignoring power draw).

Just a quick follow up: The problem is fixed now. It appears that somehow, the motherboard was unable to clear the settings of my 5GHz attempt. The HT link was stuck at 2.6 GHz and the CPU was actually throttling-up while idle (yes, the CPU clock went up instead of down when idling! crazy... ) Now I revised every setting I could see. My computer is happy again, and even I got a decent and perfectly stable 4.1GHz while the CPU is below 60ºC and VRM1 is at 70ºC

Accordingly to some sources the VRM's should whitstand temperatures of around 85°C up to 120°C but I was concerned as these numbers look crazy high... ...aaaand, the CPU temperature I took was on the thermal diode on the motherboard, the temps reported by the CPU itself are like 20-30°C lower... I've just tested this yesterday, my motherboard would report 70°C while my CPU only ~50°C I'll check the RAM first. My PSU is at the very least decent, is a Corsair HX850. I've already probed the PWR_GOOD and it's healthy. I might clean the pins just for good measure.

That sort of things happen when the card doesn't need to go on full throtle to run a game, or maybe the card hit some safety thereshold and returned to default settings. More testing could give you a hint at what's happening. Try a different game, or lower your settings a little. See what happens.

Here's the thing: I've overclocked my rig since I built it back in early 2012. It has been working fine up to 2015, but for some weird reason I thought it would be a good idea to re-visit the settings and maybe get an extra power. maybe too much. Originally I've overclocked my CPU from stock 3.1 GHz to 3.6 GHz.with no voltage increase, no weird tweaks, no nothing... just a plain increase in multiplier and DDR3 speeds through XMP, and it worked great. Then just some weeks ago, I decided to see how far could I get without destroying my computer. Somehow, I got 5.0 GHz, I ramped up the HyperTransport to 2.6 GHz, the Base clock was something like 250 MHz, the RAM was 1933 MHz I think,and I gave a boost to the VRM clocks and increased most of the voltages about 0.3v. The system worked for the first night, and I was amazed! It was almost a completely new system! the temps were high though, like 102ºC for the VRM's, and the CPU was well in the 85ºC range. The next morning, I turned on my computer, and the (almost obvious) instabilities started. I decided to go back to my old settings, but since a couple of days my computer doesn't POST unless I unplug it from the wall for a minute or so and then plug it back on. It turns on for like 3 seconds and then turns off. After the unplug-replug thing it starts behaving normally. My rig is an AMD FX8120, Asus Sabertooth 990FX, 2x Corsair Vengeance 4GB 1866MHz and GTX 660Ti, and a cheap-ass CoolIt Eco ALC Watercooler. I have already cleared the settings several times, and the problem persists.

Hiow about a quick CHKDSK?

I would get a RMA for both, but... I've voided the warranty of the portable by opening it and living outside the US (that's what an actual NV tech support guy told me) and shipping the tablet back to get a replacement would be really time consuming, and shipping rates from Honduras (incl. taxes and all that stuff) is enough to just buy another one.

If you want to go further in diagnostics, grab a multimeter (absolutely any multimeter will do) and probe the gray pin in the 24-pin connector and ground. This pin is PWR_OK (or POWER_GOOD) signal coming from the PSU, it should be 5v. I've seen Power Supplies thrown away because people thought they were bad, and it might actually be something like a loose power connector from the wall, or even the voltage selector switch, if said PSU has one. This test may not apply if you buy a really cheap PSU. -Edit: To check if the PSU has a proper self-test circuitry, you can backprobe for continuity in both the gray pin and any 5v rail. There should be at least some significant ressistance. If the resistance is near-zero, the PSU doesn't have any sort of Self-test and should stay away from it...

What a way to start in LTT Community... Anyway, I had my Old-good NVIDIA Shield Portable since I guess a year now, and usint it was pretty awesome. I used to love the thing, and played almost everything I could on it... until the screen decided to not work 5 months later. Now, I'm having some serious concerns about how durable NV products are. So, what happened was that one day I was flipping through facebook and suddenly notticed that half of the touchscreen wasn't working. I didn't mind at all, since around that time facebook app was getting updated frequently, and it was common for it to just crash or make the system unresponsive. I closed the lid, and the next time I opened it, there was no output whatsoever. I tapped the buttons and I could hear the thing alive, doing something, but the screen was DEAD. I didn't apply for the warranty, because I'm not in the USA. So, the best Idea I could come up with was disassemble it and take a look at what would be causing that, and found that the entire flex for the screen ripped appart! Obviously, that's impossible to fix in any cheap way, and I was looking like crazy for a spare screen on almost every page I could find, but nobody sells spares for the Shield portable yet. Not even iFixit, or ebay, or Amazon, or <Inser your own obscure shopping webpage here>. There's no secret about the Shield Tablet's cracked corners (which I also own, and yes, 3 of the 4 corners are cracked, one badly) and now my question is, have you ever experienced some similar issue with the Shield Portable? If you own a Shield Tablet, how awful those cracks are, if any? Here, some pretty scary pictures of my poor Shield Portable.

This is SOOOO BIG I had to join LTT community! Allright, here we go: Vessel Username: TheXDS Videos: https://www.vessel.com/videos/LCoY5zfFf https://www.vessel.com/videos/UWDhqJTDf