Search results

  1. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    That would be a good call, looks like you have an intake running already. No sense wasting those fan mounting points :thumb:
  2. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I think you're thinking is just fine, agree though a master model is a big job, hence why I went with night only and around 50 images from existing footage, it works well for headlight bloom, taillights etc. Honestly if I could have just one model covering all it would be good but potentially...
  3. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I will stick with 1, high and my usual 15-25 +real time images :thumb:
  4. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Question, does having 2 instances instead of 1 with lower + real-time images give better results? equal load on GPU/CPU maybe relatively.
  5. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    This is interesting, I really didn't pay attention to this setting, mine has been on 1 until now. What is the recommended if you have GPU with some reasonable grunt? and an old i5 processor :wtf: :lmao: and 10 cameras mix of 2mp and 4mp. I will of course experiment, as you do :thumb:
  6. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Server per camera??
  7. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I've switched Rekor back to CPU and will try running 2 instances with lower sample rate in Bi and see how it works out. I have a GPU but :facepalm:
  8. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Holy smoke, I set mine to 2 and it killed my CPU dead :lmao: poor old i5 4th Gen. I need an i7 or maybe i9 :lol:
  9. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    With a relatively low load, if it's throttling now then how much airflow does it need from case fans to stop that. I get it that these are rack mount server cards but regardless, seems like poor design. See how the new blower works out :thumb:
  10. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I like the colour. Don't understand why such an expensive card when new would be designed thermally inefficient? doesn't make sense. If they thought the cooler wasn't enough then make it bigger or add a fan(s). Seems like a classic case of make the customer fix the problems we couldn't be...
  11. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I trained my night model with essentially vehicles headlights, taillights, side view etc, works well :thumb:
  12. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Annoying isn't it, I believe it is simply due to the scene, these cameras are in pitch black with average IR, big challenge. I'm running 10 cameras so far. Try running a few cameras in the dark with IR and watch those times go up :lmao: :lmao: Daytime processing times are low, ie 60-150ms, I...
  13. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    No doubt from my brief testing, although it seems footage in zero light with just IR causes longer processing times for Deepstack using custom model, a lot longer! my times can be 3000-4000 instead of 100-200 during the day with the default model. Works well regardless, just annoys me a bit that...
  14. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Understand, I trained my own model with around 50 images and to be honest it took less time than I expected, works well. For my situation the ExDark was not much good either.
  15. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I know, my typo sorry, but no, it doesn't work :lmao: :lmao: :thumb:
  16. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Granted I have probably yet to see an actual fan failure in any I've supplied more recently - historically my money would be on a fan failure before chipset or similar. Gigabyte tend to be the quietest along with some MSi models, sadly none of the fans will be Noctua quality but those fans are...
  17. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Likely no need for custom models if you are in a well lit area, only really for darker situations potentially. See how detection rates are with the default model. Replacement fans for GPU's not so easy to come by, sadly not like case fans where you can buy a quality brand like Noctua :thumb:
  18. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Quite a few of these cards do, my 1660 is the same, fan comes on when required, normally during a heavy Deepstack event, then goes off again, card temp without fan is normally around 50 which I'm not too worried about. You could in theory run the fans all the time on low with a custom curve...
  19. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Thanks, I think a bug also, being able to use :% in 'To Confirm' has only just started working for me again on the latest release.
  20. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Well, adding :0 after any model seems to have no effect, maybe this is a GPU thing or a bug? :wtf:
  21. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    OK, am I going mad (probably :lmao:) but when I put objects:0 or exdark:0 in the custom models box it used to disable that model, what am I missing?? I am trying to force the custom models off during the day profile essentially, also the default profile off at night. Normally I just put none or...
  22. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Sounds like a genuine seller, hope all goes well :thumb:
  23. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    If you can wait then do that, maybe even look at the Tesla models, otherwise just go with the Asus :thumb:
  24. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I wouldn't lose too much sleep over EVGA vs Asus, if you can wait a while try and find another EVGA maybe.
  25. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    That would work because it will look for a custom model with the name : and it doesn't exist so defaults to the standard model. My use of semi colon has gone out the window since GPU has been running the show :wtf: for objects at least.
  26. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I though putting 'objects:0' in the models box stopped it using the default model and then used custom only, or 'objects:0,exdark' to force the custom model? that used to work on my CPU but on GPU the : command has no effect for objects and percentage, ie person:51 to only log 51% and higher.
  27. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    I hope not but possible, was like it on my CPU also and the same on the 1660
  28. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Even with the other database in theory blocked it is still just as slow for me at least. If anyone knows a way to rectify that I am all ears.
  29. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Ok, understand, let us know times if you try ExDark which is an easy one to drop in. My times for that or custom models are approximately 10-15 times longer :wow: anything up to 4000-5000ms
  30. C

    Hell Yeah! Direct Deepstack Integration - 5.4.0 - March 31, 2021

    Do you run the ExDark or any custom models in Deepstack at all?
Top Bottom