CapFrameX

@CapFrameX

Frame time capturing + analysis, OSD, sensor logging, hardware testing Contact: contact@capframex.com

capframex.com
Joined June 2019
391 Photos and videos Photos and videos

Tweets

You blocked @CapFrameX

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @CapFrameX

  1. Pinned Tweet
    Oct 24

    v1.5.7 beta with Bugfix 🚨 A critical bug was reported. v1.5.6 has a process list refresh error in some cases. If you have problems making captures install the v1.5.7 beta.

    3 replies 2 retweets 18 likes
    Undo
  2. 13 hours ago

    Step by step... 👍

    4 retweets 13 likes
    Undo
  3. 14 hours ago

    Thanks for all the great Adobe Premiere Pro tutorials on Youtube. You are my heroes. ^^

    1 reply 5 likes
    Undo
  4. 15 hours ago

    Don't get me wrong with this. I have a deep respect for what they do. It takes a lot of time and effort. But, the quality of CPUs reviews especially in the so-called Youtube tech scene is shockingly bad. #4

    3 replies 6 likes
    Show this thread
    Show this thread
    Undo
  5. 15 hours ago

    With proper CPU tests reviewers would have discovered/shown this already and the users would not be surprised. #3

    4 likes
    Show this thread
    Show this thread
    Undo
  6. 16 hours ago

    Some people might call this cherry picking but Cyberpunk 2077 is the most important title in 2020 and maby even in 2021. So how is this possible? It's not only a load problem (SMT). Let's be honest... #2

    3 replies 1 retweet 11 likes
    Show this thread
    Show this thread
    Undo
  7. 16 hours ago

    More than 1 year ago at Zen 2 launch there were some interesting test results which showed a slight difference (~7%) in gaming performance between Comet Lake and Zen 2. Now let's have a look at a test from . Cyberpunk 2077 tested at 1080p. 10900K is 35% faster! #1

    10 replies 5 retweets 25 likes
    Show this thread
    Show this thread
    Undo
  8. 16 hours ago

    I went back to %, because 0.2 for example isn't an ordinal number. The st, nd, rd, th "concept" is not universal.

    I've already changed it to "1th percentile FPS" to be more "international". #2
    Show this thread
    1 retweet 5 likes
    Undo
  9. 17 hours ago

    Of course it's called 1st and not 1th. But that's a reason why I prefer %.... ^^

    1 reply 1 like
    Undo
  10. 17 hours ago

    Here's an example from a german science website to give you a source. It's called "25%-Perzentil". #3

    2 likes
    Show this thread
    Show this thread
    Undo
  11. 17 hours ago

    I've already changed it to "1th percentile FPS" to be more "international". #2

    4 replies 7 likes
    Show this thread
    Show this thread
    Undo
  12. 17 hours ago

    Because some people are confused about the percentile notation in CapFrameX. For example, "1% percentile FPS" is unusual in anglo-saxon regions, but it's not wrong. Percentiles are 0.01 - 0.99 quantiles. 0.01 - 0.99 = 1% - 99%, because 1 per cent is just 1/100 = 0.01. #1

    3 replies 15 likes
    Show this thread
    Show this thread
    Undo
  13. Dec 15

    Similar results with CPU scene from .

    1 retweet 2 likes
    Show this thread
    Show this thread
    Undo
  14. Dec 14

    Ryzen CPUs with 4,6 or 8 cores definitely benefit from the hex edit trick.

    4 replies 19 likes
    Show this thread
    Show this thread
    Undo
  15. Dec 14

    Simulated 5600X with active hex hack.

    5 replies 6 retweets 51 likes
    Show this thread
    Show this thread
    Undo
  16. Dec 14

    Steve, can you check the results please? Maybe it's the known issue with the disappearing crowd! As a mathematician, I recommand to use another def. for the Low metrics, that is less outliers sensitive.

    Replying to @Dachsjaeger @CapFrameX
    Alex. something is fundamentally wrong with this game lol. look at this chart from the 10600K crushes the 8700K..which is basically the same CPU lol. the 9900K is...slower?.. than the 10600K. AND Loses to the 3700X how is this game even coded
    3 replies 13 likes
    Undo
  17. Dec 14

    OMG, we need AMD graphics cards to test CapFrameX. RDNA 2, RDNA 1 mobile... Any help out there for our poor souls? ^^

    2 replies 2 retweets 18 likes
    Undo
  18. Dec 14

    Solving circle references and enjoying this sound...

    1 reply
    Undo
  19. Dec 14

    Mal eine Ergänzung auf Deutsch. Warum, wird gleich deutlich. Ich bin früher in der Schule gefragt worden, was ein Beispiel für eine Einrichtung sei, die unabhängige Produktests macht. Antwort: Stiftung Warentest. Versucht mal kostenlose Tests von den zu bekommen? ^^

    1 reply 8 likes
    Show this thread
    Show this thread
    Undo
  20. Dec 14

    If we really want to have an independent tech press we need to go away from free samples for reviewers who are advertising-sponsored. Reviewers would buy the hardware at launch day like everyone else. And we consumers would have to pay for the tests.

    11 replies 3 retweets 37 likes
    Show this thread
    Show this thread
    Undo
  21. Retweeted
    Dec 13

    As said, Nvidia's Ampere picture was close to reality but again, close doesn't mean it's precise. My annotations are also rough on the real die shot, there isn't a clean parting line on some structures, I just made a cut there. I/O (NVLink + PCIe4) is trickier...

    1 reply 10 retweets 62 likes
    Show this thread
    Show this thread
    Undo

@CapFrameX hasn't Tweeted yet.

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·