Jump to content

mvitkun

Member
  • Posts

    4,264
  • Joined

  • Last visited

Reputation Activity

  1. Agree
    mvitkun reacted to The Benjamins in Nvidia thinks "Pascal is just unbeatable" and already decided not to move Volta ahead   
    Bad word use on my part but I meant, hardcore CS:GO and dota 2 players. the people that care about 144hz gaming. would also play other games.
  2. Like
    mvitkun reacted to AnonymousGuy in Gamers Nexus tests Noctua fan rumors.   
    Someone needs to ask Noctua then what the fuck the purpose of all their fins and grooves and shit are then if they can be fucked up badly and really not make much of a difference.
     
    Also, he really should have done a frequency spectrum.  Because I suspect ^ probably translates to reducing noise at specific frequencies that may not contribute much to total dB.
  3. Informative
    mvitkun reacted to leadeater in [Debunked] BSD team discovered a new hardware issue with Ryzen   
    How is this happening when SMT is disabled when the issue been described is with SMT itself? The other mention of system stability seems to be a different problem.
     
    This seems like a good explanation to the cause of some of the issues:
    https://www.phoronix.com/forums/forum/hardware/processors-memory/955368-some-ryzen-linux-users-are-facing-issues-with-heavy-compilation-loads/page7
    Sounds like GCC needs some Zen optimizations and a lot of software recompiled and update versions pushed out.
     
    Also there's an update to the Segfaults issue that points to potential improper testing.
     https://www.phoronix.com/scan.php?page=article&item=ryzen-segv-continues&num=1
  4. Agree
    mvitkun reacted to leadeater in [Debunked] BSD team discovered a new hardware issue with Ryzen   
    @zMeul Instead of marking my post funny you could actually try an answer my very simple question. It didn't need to be an argument at all, this issue is talking about SMT then you said it exists when SMT is disabled and I'd like to know how that is possible, not a hard question. Especially when the original source says the bug does not happen when the cpu-bound loop is on a different core than the core doing the IRETQ, the only way they can be run on the same core is with SMT.
     
    Marking posts as funny is rather childish.
  5. Funny
    mvitkun reacted to AnonymousGuy in RX VEGA Prices Confirmed   
    This is the most apt youtube video to describe the situation:
     
    "Don't tell me this is Zune bad"
  6. Funny
    mvitkun reacted to Valentyn in AMD Won't be livestreaming Siggraph and Vega's launch   
    AMD  when people ask about Vega performance, and wanting them to Livestream Siggraph and demo live.


  7. Informative
    mvitkun reacted to Valentyn in AMD Won't be livestreaming Siggraph and Vega's launch   
    Hmm top card shows 8+6 pin physical.
     

     
    Bottom has 8+8 Physical; even though wiring is 8+6. Hmmmm
     

  8. Like
  9. Informative
    mvitkun got a reaction from NvidiaIntelAMDLoveTriangle in Adobe pulls the plug on Flash. RIP Flash 2000-2020   
    Just use this site to find all the versions of the video, copy the desired url, open VLC player, media -> open network stream, paste the url, play.  
     
    This works up to 720p, beyond that youtube stores audio and video separately so you'll have to do this:  
    copy the desired video and audio urls, open VLC player, media -> open network stream, paste the video url, show more options, play another media synchronously, paste the audio url, hit play.
     
     
    EDIT:Just wanted to note that the MP4 stream takes longer to load than the webm stream.
     
     
    You could also use youtube-dl
  10. Informative
    mvitkun got a reaction from Jito463 in Adobe pulls the plug on Flash. RIP Flash 2000-2020   
    Just use this site to find all the versions of the video, copy the desired url, open VLC player, media -> open network stream, paste the url, play.  
     
    This works up to 720p, beyond that youtube stores audio and video separately so you'll have to do this:  
    copy the desired video and audio urls, open VLC player, media -> open network stream, paste the video url, show more options, play another media synchronously, paste the audio url, hit play.
     
     
    EDIT:Just wanted to note that the MP4 stream takes longer to load than the webm stream.
     
     
    You could also use youtube-dl
  11. Like
    mvitkun reacted to Ginger_ in Worse Noctua fan quality after moving production to China   
    I work for a small oem, that we sub contract out a big part of our final product to a company in China. This is exactly what we face. We approve a design, then slowly the quality degrades.
     
    Hell one time he had a mould changed because we discovered a lot issue with a power connector coming loose from the battery, our fault because it was our design, so they said they tossed all their extra stock of the old bad pieces and charged us for it. Couple months later we start seeing a device here, device there that they put the old piece in. Plus probably a dozen times they've changed plastics, thinned certain parts etc.. 
     
    Apparently from talking to other companies this is quite common from Chinese companies. I can only imagine noctua is going to face the same thing as well and if they don't catch them, as customers were screwed. 
  12. Informative
    mvitkun got a reaction from LAwLz in Cybenetics releases psu certification for mining   
    By use do you mean send their power supplies to Cybenetics for certification, use the certification on their marketing material, or something else?
     
    They currently have 79 power supplies on their efficiency database and 76 on their noise level database, mostly from relevant power supply vendors.
     
    I'm not sure how many of them actually advertise the certification, most of them don't seem to have updated their product pages to include it but Seasonic displays it prominently alongside the 80+ Titanium certification.
     
  13. Informative
    mvitkun got a reaction from LAwLz in Cybenetics releases psu certification for mining   
    Yeah I don't expect to see a filter for it until it becomes much more widely adopted, assuming that it ever does.
    EDIT: I checked their news page and apparently there's one etailer that has a page with all the power supplies they carry which Cybenetics have tested.
     
     
    Oh and I just checked the rest of the companies on their database, Silverstone displays it as well but it seems that nobody else does.
     
  14. Agree
    mvitkun reacted to Jito463 in Intel undermines AMD's new EPYC Server Lineup performance in its Official Xeon SP Slide Deck   
    I know people have already pointed out how Ryzen is way underclocked for this comparison, but I also find it interesting that they compare a 6 memory channel CPU against a 2 channel CPU.  Memory bandwidth is going to be severely limited on Ryzen at that point, compared to the Xeon.
    I agree.  The biggest surge in Epyc sales will occur - at minimum - 1 to 2 years from now, assuming everything goes as planned, and there's no major hiccups.  There will be some initial sales (probably for testing/validation purposes primarily), but most companies aren't going to jump right into a new architecture immediately after launch.
     
    Shouldn't that be finale?
  15. Agree
    mvitkun reacted to JoostinOnline in Intel undermines AMD's new EPYC Server Lineup performance in its Official Xeon SP Slide Deck   
    I burst out laughing when I saw that they were comparing it to an underclocked 1800X.  The Xeon wasn't even that much better!
     
    Sorry, but no you don't, unless you mean an older generation that has been tested.  They're probably fine, but you can't know there won't be any problems.  People assumed they were getting a 100% stable product with the 7700K, but those get random heat spikes.  This isn't a dig at Intel, it applies to everything.  You can't assume there won't be any problems with something new.
     
    I totally understand not wanting to switch to AMD right away.  Trust needs to be earned back.

    That's immediately what I thought of when they said "glued on".
     
  16. Like
    mvitkun reacted to LAwLz in Tech companies to 'nerf' online services on July 12th in a bid to keep net neutrality   
    Except your understanding of what net neutrality is was fundamentally flawed so the source material you linked was completely irrelevant.
    Here is the thread for those of you who are interested.
     
    Edit:
    People who are against net neutrality don't have any idea what they are talking about most of the time.
    Saying that we should not have net neutrality because it didn't help is like saying we should make murder legal because "the law clearly doesn't help since people are still being murdered". Having broadband classified under title II is the first step in making things better. In fact, when the FTC tried to make things better without Title II classifications they were sued by Verizon, and Verizon won.
     
    A lot of people are annoyed at ISPs for things that Net Neutrality won't address, so people go "well they are still shit so it didn't matter", when it might have done something. Let me tell you, as someone who briefly worked as a first line support at the second largest ISPs in Sweden, 90% of people who call in and complain to their ISP are wrong, and yes I am sure a large portion of people here (maybe even in this thread) are wrong about their ISPs. Just because your Internet connection is slow does not mean you are being throttled. There is certainly throttling going on in some situations, but I honestly would not be surprised if 50% of people on this forum who complain about throttling are actually not being throttled.
     
    Net Neutrality laws won't solve data caps, congestion, poor infrastructure or anything of that. But those are separate issues compared to traffic prioritization at ISPs. Throwing out net neutrality because it doesn't solve the aforementioned things is a bad idea because it will only make things worse, without solving any of the issues.
     
    Anyway, I am not in the US so if you want to ruin your terrible Internet situation even further then go ahead.
    But before you start arguing for making everything worse, maybe you should consider why you want to get rid of Net Neutrality and how it would make things better?
    And I don't mean "net neutrality didn't do anything so we should get rid of it". The question I want people to think about is "if Title II classification was removed, how would things get better". If you can't think of a good reason why things would be better without Title II classification, then you need to think about why you are against it in the first place.
  17. Agree
    mvitkun reacted to Oshino Shinobu in [Update] China tells carriers to block access to personal VPNs by February 1, 2018, [Bloomberg is reporting erroneously]   
    Theresa May (our Prime Minister) has made it clear she has intentions to regulate and restrict what can be said, posted and seen on the internet, also making it clear that she will change human rights laws to accomplish that and make the UK "the global leader in the regulation of the use of personal data and the internet". 
     
    She wants an outright ban on full encryption, basically a scrapping of the rights to privacy and freedom of speech and regulation on what is and isn't acceptable on the internet (the UK Government seems to have an irrational obsession with what type of porn people are watching). 
     
    Basically, our government looks up to China in the way they handle internet access. Though, to become the "global leader" in internet regulation, that would mean surpassing North Korea as well. Say hello to Kim-Jong-May and her field of wheat and oppression. 
     
    http://www.independent.co.uk/life-style/gadgets-and-tech/news/theresa-may-internet-conservatives-government-a7744176.html
     
    EDIT: Of course, there has been no actual information on how they plan to really enforce or accomplish this, or how they're going to either find the massive amount of funds it would take or get companies to pay for it. I kind of hope big companies just pull services entirely from the UK, N.Ireland joins back up with the Republic, then I can move over there and leave the UK to sit in the mess that it has made for itself. 
  18. Like
    mvitkun reacted to mayhems in Nano Graphene Water Cooling Coolant In Development   
    Hello All.
     
    Yes we have been working with CPI for a very long time so the message starter on OCN is a little deceiving and ill correct that. We have worked with CPI in the past to make several coolants more stable and the knowledge we have gained from that is allowing us to produce a new coolant from the ground up. The new coolant is a Graphene based type product (cannot go into much detail about it).

    What are its advantages so far  and why make it?
     
    1) Its is Chemically inert to PH imbalances.
    2) Uses a very limited amount of chemicals.
    3) Possibly up to 20 to 30% better cooling than water alone (tested in lab environment not real world as of yet).
    4) 1 nm thick (pastel is 40nm and aurora 40 to 100nm)
    5) Completely new tech never been done before.
    6) Uses tech that lubricates pumps and o-rings on the fly with out wetting them.
    7) Will prolong the life of your equipment.
    8) Should out last the life of your PC. (TBC)
    9) Initially will cost a little more to start with but should come down in price as and if we do go into full production, 
     
    Why do it ... why not :). 
     
    Initially it will not be coloured as this effects its raw performance.
     
    Mick
  19. Informative
    mvitkun got a reaction from DocShay in Nano Graphene Water Cooling Coolant In Development   
    You might want to quote what he said in the OP for people who don't read the entire thread.
  20. Funny
  21. Agree
    mvitkun reacted to SpaceGhostC2C in [UPDATED] Der8auer States Some X299 Low End boards Have Bad VRM Coolers   
    So: we have boards with the pcie and dimm slots for Skylake-X and the power delivery for KabyLake-X. 
    I guess it's the result of the marvelous idea of having one 4 to 18 cores, gaming to server platform...
     
    I think that's what may be happening: manufacturers don't want to overspend on the lower X299 end, with CPUs that could basically take motherboards copy-pasted from Z270. However, they cannot release separate boards, as omen requires support for the full lineup on all X299 boards. 
    Hence, in order to compete, manufacturers have to release lower cost boards for lower cost "X299" builders, but they cannot cut down on the truly useless stuff for "CPU formerly known as 7700K", so they cheap out on what is left at their discretion, and "officially support" any socket 2066 CPU. If der8auer's findings are confirmed, the unofficial (or maybe "updated") support list for these boards will be considerably shorter. 
     
     
    In other words: the trakeout isn't "X299 will catch fire, avoid at all costs", but rather that the "one platform, 300 CPUs" isn't really so (hence why he focuses on these boards). 
  22. Agree
    mvitkun reacted to WolfDeville in MSI Afterburner adds graph plots directly to the OSD - BETA build   
    Title nearly gave me a heart attack... Please correct "ads" to 'adds'
  23. Agree
    mvitkun reacted to vanished in Nvidia Volta Won't Use HBM2, But GDDR5X   
    Is this surprising?  Really?  When did we hear they were going to use HBM?
  24. Agree
    mvitkun reacted to Sauron in Windows XP patched to avert new outbreaks from three more NSA exploits   
    Despite the obvious problems with issuing a one-off update to xp, there are hospitals with medical equipment that only works with windows xp. That equipment may have cost hundreds of thousands if not millions of $ and governments are not always as keen on spending money on health as they are on military and other garbage. Preventing stuff like WCry from infecting hospital machines could potentially save lives, and since these exploits are now public knowledge ignoring xp might be considered criminal negligence. It's easy to point at Apple and say "see, they don't support 16 year old discontinued products anymore" when apple has never had a presence in the mission critical life saving apparatus market.
  25. Like
    mvitkun reacted to LAwLz in Apple Abandons JPEG In Favor of HEIF/HEVC   
    VP9 is nice, but it got some issues.
    1) The encoder and decoder were not very good until fairly recently. A recently update to the encoder improved the encoding speed by about 60% in some tests.
    2) Support was not as widely spread. Windows still does not support VP9 outside of Edge, which only added support very recently. Windows 10 has supported HEVC since day one. Since AV1 is backed up basically all big tech companies, except Apple, wide adoption is 99.9% guaranteed.
    3) The patent situation was/is somewhat unclear. MPEG-LA (which are a completely different group than MPEG) threatened to sue companies using VP8 because it infringed on H.264 patents. That threat did not lead anywhere, but because of it (and similar threats made towards for example Theora), many companies felt like it was better safe than sorry and went the "we'll just pay and be sure that we're safe" route.
    4) VP9 is "current-gen" compression. AV1 is going to be next-gen. VP9 currently provides about the same image quality per bit as HEVC (H.265). AV1 is aiming to (and I am fairly sure it already does) provide the same image quality at even smaller file sizes (or better quality at the same file size) than HEVC. It's going to go up against H.266 or whatever it will be called.
     
    Since the licensing situation was getting out of hand, several companies started developing their own codecs meant to replace HEVC. Google started developing VP10. Cisco started developing Thor. Mozilla/Xiph started developing Daala.
    Then they all decided that it would be better if everyone just started developing a single codec instead of three different ones. So they formed the Alliance for Open Media (AOMedia) along with a lot of other companies. A huge amount of work is being done by AOMedia to make sure that nobody will get in trouble for using it, just to avoid the current uncertainty that surrounds VP8 and by extension, VP9.
    The license for AV1 is (currently) really good.
     
    Some highlights:
    1) It's open source and does not involve any licensing fees.
    2) The license is short and fairly easy to read (unlike for example Opus which has two separate licenses which you need to read).
    3) It uses the same license for both the (reference) encoder and decoder. When you buy a license for HEVC, you will get a completely different license depending on if your license covers the encoder or decoder. So if you want to do both, you will need two separate licenses that you need to pay for and make sure you do not violate.
    4) It includes a very broad and strict defensive termination clause, which means that if you sue anyone for using the AV1 codec, you lose the right to all patents involving AV1. That means that if Netflix decided to sue you for using AV1, then Netflix will lose the right to use all the patents Microsoft, Mozilla, Google, Amazon and everyone else have contributed. So basically, suing anyone over AV1 is suicide because it's like starting a fight against the entire tech industry.
    5) The license passes the so called Debian desert island test. In practice what it boils down to is that the license is embedded into the actual product, which means that anyone using AV1 has been granted permission to use it. You do not need to sign anything, or ask for permission. Anyone can pick up AV1 and use it, and by doing so they are granted a license.
×