redteam4ever

And where did I mention AM4? I wanted to write AM4 originally, but when I read the article, I've seen they were talking about the architecture. I never mentioned AM4 in my post. And thanks, but I know the difference between socket and architecture.

Quote from this article: I have mixed opinions on abandoning x86. Since learning more about the instructions (and memory management in x86 and x86-64), I think it can kind of set us back, but I'm also afraid what it will do to the whole software ecosystem. Also, to clarify, I didn't mean that Zen 1 is going to be here for 4 years, I meant also the Zen derivatives (as said in the article which I researched before correcting AM4 to Zen when I wrote my reply).

They stated that they want to be 2 generations ahead when they announce a new one. So they are already working on Zen 2 and Zen 3. They also stated that Zen will stick around for 4 years. We'll see, but I don't think that they will let themselves fall behind again, especially when Radeon is not being competitive in high-end. They probably also hired new people during Zen development, so those are the ones that will help. I just hope that x86 will still be here after four years.

Well, it's not like he designed the CPU himself. In Ryzen Tech Talk, Suzanne Plummers (Sr. Director Design Engineering) and Michael Clark (Sr. Fellow Design Engineering) briefly touched on FX, saying (about performance) that "they were not getting where they wanted to be" (3:18). Clark's LinkedIn profile says that he has worked at AMD since 1993. There's a lot of engineers that designed Zen and a lot of them were probably also on FX design team, so that's that.

I don't see any problem with that. Just right click the drive in Computer and if the formatting option is not greyed out, it could be done. If it's greyed out, it's write protected and nothing can be done in that case. But I wouldn't do it, I doubt that Microsoft opted for quality and it will probably perform terribly (I bet on slow write speed).

No, we have DSL everywhere, and the observed ping was only for DSL (recently I've connected some people on my street to fiber, I haven't seen more than 10ms on fiber). My big city comment was more about DSLAM density and wiring quality as it's more probable that wiring it's newer than in rural areas where the cost of rennovation per connection point would be too high.

It would be helpful to provide links to IKEA's website as your description is not good enough. As for adding personality, I would add some glow. IKEA sells LED strips, but they have separate controller and can't be synchronized with anything, so maybe some other LED strips with HUE+ or something. I don't know what you mean by hiding the PC behind the drawer, but Linus had his HTPC inside some furniture and this video might help you if that's your case.

I see that you know more than I do. By abnormally high, I mean that I have never seen that in any place that I was helping to get the internet from Telekom going. I don't work at Telekom, but I helped a lot of friends that are non-techies when they've got the self-installation kit from them because it's still quite confusing to non-techies. I always do a speed test when I'm done and I've never seen more than 30ms anywhere (but I live in a relatively big city). If it's as easy as changing the configuration, then the solution is very simple - contact Telekom at 0800123777 and complain about high latency. Keep in mind that they will ask you about the contract details to verify your identity. Also they may provide additional troubleshooting steps.

Yeah, I know. But at least they replaced most of their DSLAMs. Telekom is also a mobile provider here. They are currently the leader in LTE. It used to be Eurotel, from 2004 T-Mobile (owned by Telekom) and then they merged with Telekom in 2010 (the mobile network quality is the Eurotel/T-Mobile legacy). Telekom has the best LTE coverage and speeds (based on P3 Communication independent testing), but the prices are high and data limits are low. And with the new EU roaming regulations I don't see it changing soon (but I don't think that regulation is bad either). Basically, you can get 15GB of data for 30€/month. So, there's that. Bottom line: They won't move to 4G for homes. Slovakia is a small country with a lot of mountains and it would cost too much.

Well, it depends on location. Telekom has the only nationwide network here. Other providers only rent Telekom's network capacity. At least they implemented VDSL2 on metallic and finally started to move their endpoints to fiber. It's slowly getting better. Some smaller ISPs have metropolitan fiber networks. Also Orange and SWAN have some fiber coverage, but mainly in places with dense population (read: where these magical buildings are).

Since you've tried running a cable from the router itself and that didn't help, there is nothing that can be done. You should test other servers like Google DNS servers (8.8.8.8) - my ping is 11-13ms and I have internet from a local WiFi provider (I'm connected through their WiFi and then through my WiFi router at home). Your latency is abnormally high and there could be a problem somewhere else possibly (Telekom's infrastructure). The other thing I'm thinking of right now is that you can try contacting Telekom if you find out that your latency is high on other servers (like Google DNS). They have a telephone number for reporting malfunctions (0800123777). Keep in mind that you will need your parents for that as they will ask details about the contract as name, address to verify caller's identity. I know something about Telekom's infrastructure. It's old and they are squeezing every last bit of bandwidth from their old cabling and network gear. In a lot of places they have 20+ year cables. I'm very lucky that they reconstructed my street and connected everyone on fiber this year. But before that, they couldn't even connect us to the internet and few lucky people on the street with internet from them were having a lot of connection issues, instability and low speeds. If you want a better connection you can try entering your address at www.porovnat.sk. It will show you all available ISPs at your address.

Exactly. This is Gigabyte's BIOS, recently I've read a review on their new Ryzen ITX board and the author complained about clunkiness of the UI: You can read the rest here...

They talked aboutFloatplane a lot in last two afterparties, you can still wtach them on Twitch: Last afterparty - July 8, 2017 (1:39:32) Afterparty most relevant to original post - Jun 17, 2017 (0:22:26)

German deals (applies also for other EU countries): Samsung 850 EVO 500GB 157.99€ | -15% 134.90€ https://www.amazon.de/dp/B00P73B1E4/ref=gbdp_vlo_9652a38f_B00P73B1E4?_encoding=UTF8&smid=A3JWKAKR8XB7XF Asus Zenbook UX3410UQ GV077T 1299.00€ | -20% 1039.00€ https://www.amazon.de/Zenbook-UX3410UQ-GV077T-Notebook-i7-7500U-GeForce/dp/B06VVZ17VN/ref=lp_13756478031_1_2?s=computers&ie=UTF8&qid=1499784825&sr=1-2 Asus Zenbook UX330UA FC080T 1099.00€ | -27% 799.00€ https://www.amazon.de/UX330UA-FC079T-Notebook-i7-7500U-Festplatte-Arbeitsspeicher/dp/B01L312IHC/ref=lp_13756478031_1_3?s=computers&ie=UTF8&qid=1499784825&sr=1-3 Asus Zenbook UX510UW-CN058T 1299.00€ | -25% 969.00€ https://www.amazon.de/Zenbook-UX510UW-CN058T-Notebook-Intel-Nvidia/dp/B06XFF5KQ3/ref=lp_13756478031_1_11?s=computers&ie=UTF8&qid=1499784825&sr=1-11 HP Envy 15 – AS102NG 1199.00€ | -21% 949.00€ https://www.amazon.de/HP-15-as105ng-Ultra-Notebook-Laptop/dp/B01M9CPTH2/ref=lp_13756468031_1_1?s=computers&ie=UTF8&qid=1499785497&sr=1-1 Sony MDR-1A 144.90€ | -21% 115.00€ https://www.amazon.de/Sony-MDR-1A-Resolution-Kopfhörer-Definition-Treibereinheiten/dp/B00N3WWG4U/ref=gbps_tit_m-2_3de6_70eeb037?smid=A3JWKAKR8XB7XF&pf_rd_p=905302be-e596-4599-a26a-7c25b11c3de6&pf_rd_s=merchandised-search-2&pf_rd_t=101&pf_rd_i=10407967031&pf_rd_m=A3JWKAKR8XB7XF&pf_rd_r=7Y56E7F01Z3Y3Y66PV4Q

Crowdsource - app by Google for improving Google Translate, handwriting recognition, Google Maps and their other services. It's the only app I use to kill time.

You should try installing the backplate to see if the metal pins really touch it. If they do, you could just wrap the backplate in electrical tape. Just make sure that the backplate wouldn't introduce too much force on those pins, since that could damage the tape. I also managed to find a post on an MSI forum that has listed this board and cooler combo, so it looks like someone is using it without problems.

I have reviewed the photo and I think that cable is the issue. If we look at the specification of 6-pin PCI-E power cable, there should be 3 yellow wires and 3 black wires. Yellow wires carry 12V and black wires are ground with zero potential. Top connector is missing one 12V wire which is chained to the second connector at the bottom. Result is that GPU is not getting enough power. However, more power is being pulled through other wires and PCI-E connector, so it runs for a while. With growing GPU utilization, it stops being enough and GPU becomes unresponsive. If the cable on the left from the damaged one was missing, your system wouldn't POST. That is one of the two sense pins that detects if the cable is plugged in. I don't know what happened to the PSU, but you should RMA it or buy a new one.

A little messed up? It might be the problem... Improper power delivery can cause a lot of issues on a GPU. Can you describe it? Or take a photo?

This might by just a round-off error if the number is being computed somehow and not taken from EDID.

Thank you very much for the nice words Just trying to help when I can. So, I looked at the log - I don't think that CUSTOM ERROR is actually an error, it's more of a log entry with memory dump - error codes are 0x0 = 0 = successful operation (usually). I didn't even find the corresponding kernel-power event. And now to many VIDEO_SCHEDULER_INTERNAL_ERRORs - bugcheck parameter 1 is weird in all of them. MSDN page is documenting values 1 to 4, yours is 5. That is weird and I don't get how that happened. Usually bug codes are added to code manually and then documented... I found a very neat NVIDIA forum post about driver resets that I talked about in some of my previous posts - when your screen goes black, Windows is trying to restore the driver. This quote is telling us what causes the reboot: "If the operating system detects that six or more GPU hangs and subsequent recoveries occur within 1 minute, the operating system bug-checks the computer on the next GPU hang." It also confirms what I thought - Windows can't reset the driver to working state (it goes to RESET-ERROR loop), so it reboots and creates the bugcheck event. There are some troubleshooting steps in the post and I'm not discouraging you from going over them, but I think that this is a problem of the GPU. You are getting almost random graphics-related errors at this time and when you fix one, another one appears. What we can eliminate from that list? Incorrect memory timings or voltages - memtest would fail. Insufficient/problematic PSU - explained in the next paragraph. Corrupt driver install - I don't think so, you already reinstalled it like million times. Overheating - events are happening rapidly and in random intervals. A badly written driver or piece of software, but this is an unlikely cause in most cases - Is NVIDIA trying to confess here? Asking too much from GPU - you are not asking too much. Driver conflicts - we can never rule this out, but it's almost impossible to diagnose. These are the ones to go over: Unstable overclocks (GPU or CPU) - just try dialing it back to stock. Incorrect MB voltages (generally NB/SB) - ^^^^^ Faulty graphics card - this is the winner IMO. Testing the GPU in another PC is a good advice, you can also try another GPU in yours. But stick to NVIDIA so you won't have to install other drivers and make things more complicated. I also suggested retightening all internal connections mainly to force you to check the GPU power cable, but I don't think this has anything with the PSU, you would be getting harsh shutdowns due to fluctuations in voltage and no BugChecks. I'll be damned if this is a system problem, you can try running system file check - open command line and type (without quotes): "sfc /scannow". Hit enter, let it run, it should be complete in an hour. One interesting thing that I noticed in the logs is that AMD SMBus driver installation si always failing, try to find the driver and install it manually if you want, but I think that we have other bigger problems

I interpreted the "screen goes black" wrong, this result is no surprise then. Please, post the Event Viewer log and I will dig through it to find the cause. Now that I have a clear picture of what you meant by screen going black, it looks like the GPU driver stops responding because GPU is not responding. GPU then stops pushing new frames, screen goes black while Windows is trying to restore the driver. Event log will tell us if this is the case.

OK, how is memtest doing? If it still says zero errors, memory is OK. It would've found some error by now with your frequency of black screens. You might also try to retighten every physical connection in your PC. Then I recommend checking out the Event Viewer. If there are graphics (or graphics driver) related problems, I would just RMA the card. Black screen might also mean that driver stopped responding. Windows can resolve this and resume the driver. If it reboots it probably means that Windows can't resume the driver because hardware is not responding meaning the GPU is defective. There is actually countless number of possibilities so it could be just about anything. If you have any question, ask now, it's 4AM here and I'm slowly going to sleep. I will be near PC again in ~18 hours since now.