• 1 Post
  • 8 Comments
Joined 8 months ago
cake
Cake day: April 5th, 2024

help-circle
  • ChiralOPtoBuildapc@lemmy.worldSemi-Portable Engineering Rig
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    8 months ago

    Great point, it’s not so clear cut of a lead. 14900K wins in price, but I haven’t factored in the cost of the mobo that might even out the total cost.

    A case like Silverstone Alta G1M would let me fit a 360mm water cooler in there, but the build is starting to get larger and larger :P A lower TDP would need smaller coolers and fit smaller cases like you said.

    I can fabricate a handle for it, and I’m considering perhaps smaller cases that I could stuff a 360mm cooler into where a GPU would be. e.g. Mechanic Master C30pro (aside from the glass…). As long as the case can handle a long enough GPU, it would fit a 360mm radiator (~400mm long) and I could bolt it to the side panel and cut some ventilation. Sadly many of the cases out there seem to be tempered glass side panels… can’t cut/drill that!



  • Great website…

    Wouldn’t a mobo advertise that its VRMs would support sustained 253W as a selling point? It was my understanding that AMD couldn’t touch Intel’s single-threaded performance. ST performance isn’t the be-all-end-all performance metric, but it seems relatively important for my use.

    You’ve got me looking at Threadrippers now and… 350W TDP???




  • Yeah, that’s kinda the typical approach. I’m hoping I can piece together more performance and less noise for the same price. Kind of a waste of a battery and screen if I’m just going to plug it into a station every time… The comparable laptops do alright but they get damn hot and they cost $5k


  • Yes! I’ve even tried it, but the issue I ran into was that my Starlink home connection is… spotty (I really need to move dishy up higher to get fewer obstructions). This results in network dropouts every couple of minutes which is suuuuper frustrating when you’re doing 3D modeling work. Also, working over a remote connection seemed pretty clunky at times with the 3D model manipulation, def not as smooth as local.