RISC-V Motherboard For Framework Laptop Pricing Starts At $368 In Early Access

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Quackdoc
    replied
    Originally posted by L_A_G View Post

    Miserable little misers who want you to sell stuff at below break even are rarely a market worth pursuing. This is a developer board and that's never going to be a big seller no matter how well you price it. If you price it low enough you also end up with the problem of miserly users buying it and then complaining that its too buggy (duh?) and can't run regular x86 executables and is too slow.
    the DC roma I is the exact same soc, the same gpu the same most things, in a different layout. The DC Roma I is 300usd. A barebones framework config is what, 1000usd? Pair that with this and now you are looking at 1200/1300 usd.

    There is a devboard tax, and then there is just being dumb. You either take your 1000+ USD laptop and make it not usable for daily driving, or you spend 300 dollars and get a second, perfectly usable laptop that is going to do literally everything your 1200usd kit out can do minus some things like a "really high quality screen"

    You don't get speed increases, you don't get a different set of bugs unless you are hitting hardware ones, you still can't run reguler x86 executables etc.

    Leave a comment:


  • L_A_G
    replied
    Originally posted by Quackdoc View Post
    They still have to convince people to buy it, This is essentially pissing away money instead of just getting the DC Roma I as now you are down a laptop. if they want to charge this much, it should at least be daily driveable in general​
    Miserable little misers who want you to sell stuff at below break even are rarely a market worth pursuing. This is a developer board and that's never going to be a big seller no matter how well you price it. If you price it low enough you also end up with the problem of miserly users buying it and then complaining that its too buggy (duh?) and can't run regular x86 executables and is too slow.

    Leave a comment:


  • Quackdoc
    replied
    Originally posted by niceride View Post
    Yeah I am not okay with being "quoted" and snipped literally the whole content of my post. Get gone, creep.
    what on earth are you talking about? you posted a near essay, anyone can click the >> button beside the Originally posted by <USERNAME HERE> for context.

    "snipping" is a courtesy thing to prevent threads from getting super long, if you don't understand that, you don't belong on a forum, but considering your kneejerk response was "creep" I shouldn't be surprised you don't understand that...

    Leave a comment:


  • niceride
    replied
    Originally posted by Quackdoc View Post

    the issue is that the price the board is selling for, vs what it gives, doesn't make sense. Especially considering the form factor, you have to take your thousand dollar laptop, take the guts out, and put a barely usable, nearly 400 dollar device into it. It would be one thing if this was some chromebook esq cost laptop, but it's not, framework devices have a high entry fee, and then the motherboard has an insanely high entry fee especially considering the cost of other things in the ecosystem, and I get that the cost might make sense from a development point of view.

    but for someone actually looking into a "riscv laptop" the specs just don't make sense. It's not something usable for daily driving, even if your distro did ship with compiler optimized bins, and now they have to replace, in many cases their likely daily driver. It should at least be a soc that hits the minimum bar for a daily driveable system.
    Yeah I am not okay with being "quoted" and snipped literally the whole content of my post. Get gone, creep.

    Leave a comment:


  • Quackdoc
    replied
    Originally posted by brucehoult View Post
    You're lucky to get a distro -- other than the manufacturer's one -- that enables RVV in the kernel. e.g. Milk-V's ones for the Duo, Meles, Pioneer, Sipeed's images for the LicheePi 4A, and Bianbu for all the SpacemiT boards (LicheePi 3A, Jupiter, BPI-F3 etc).

    With that you can build your own apps using vector.

    I don't expect to see distros with libraries and especially packages compiled for RVV before e.g. Ubuntu 26.04 and I think they've promised 25.10 as well.

    I also expect it will be a long time before Geekbench update their code to use RVV when it is present. There are many reasons why Geekbench is very problematic to use to compare different ISAs, but that is going to continue to be a major one with RISC-V numbers for years to come.

    The EIC7700 is a fast chip (by current RISC-V standards) for running generic code on e.g. building software, but it's just RV64GC, not even RVA22.
    I see, I do plan on getting a machine with rvv, I may just rebuild most of the base archlinux packages with rvv support so folk can use that to get a more optimized experience. I also plan on looking into what would be needed to get blissOS running on riscv, but the android stuff seems fairly immature still.

    Leave a comment:


  • brucehoult
    replied
    Originally posted by Quackdoc View Post
    RVV will be nice for sure, but if nothing is compiling apps for it it's going to be rough, do we know if any distros are shipping binaries that have rvv enabled?
    You're lucky to get a distro -- other than the manufacturer's one -- that enables RVV in the kernel. e.g. Milk-V's ones for the Duo, Meles, Pioneer, Sipeed's images for the LicheePi 4A, and Bianbu for all the SpacemiT boards (LicheePi 3A, Jupiter, BPI-F3 etc).

    With that you can build your own apps using vector.

    I don't expect to see distros with libraries and especially packages compiled for RVV before e.g. Ubuntu 26.04 and I think they've promised 25.10 as well.

    I also expect it will be a long time before Geekbench update their code to use RVV when it is present. There are many reasons why Geekbench is very problematic to use to compare different ISAs, but that is going to continue to be a major one with RISC-V numbers for years to come.

    The EIC7700 is a fast chip (by current RISC-V standards) for running generic code on e.g. building software, but it's just RV64GC, not even RVA22.

    Leave a comment:


  • Quackdoc
    replied
    Originally posted by brucehoult View Post

    The next generation chips (except the EIC7700) have RVV 1.0, which is better than NEON.

    The DC-Roma laptop I mentioned for $299 has the same SoC as this. The DC-Roma II has an 8 core SoC with RVV.

    We can probably expect a 2nd gen RISC-V Framework mainboard with a newer CPU, such as the SpecemiT in the DC-Roma II, within six months, once this board has shaken out the integration bugs.
    yeah im pretty excited for the eic7700, the 256flop per clock gpu is also going to be a massive improvement depending on it's clock speed. The spacemi isn't a great value in this regard as the BXE-2-32​ it uses is fairly slow, but the bump in cpu perf will be nice for sure.

    RVV will be nice for sure, but if nothing is compiling apps for it it's going to be rough, do we know if any distros are shipping binaries that have rvv enabled?

    Leave a comment:


  • brucehoult
    replied
    Originally posted by Quackdoc View Post
    unfortunately neon is a fairly massive killer.
    The next generation chips (except the EIC7700) have RVV 1.0, which is better than NEON.

    The DC-Roma laptop I mentioned for $299 has the same SoC as this. The DC-Roma II has an 8 core SoC with RVV.

    We can probably expect a 2nd gen RISC-V Framework mainboard with a newer CPU, such as the SpecemiT in the DC-Roma II, within six months, once this board has shaken out the integration bugs.
    Last edited by brucehoult; 15 November 2024, 09:01 PM.

    Leave a comment:


  • Quackdoc
    replied
    Originally posted by L_A_G View Post

    We're not talking about essentials like bread or rice here so no prices are ever "justified" no matter how high or low. Also, that price is for a raspberry pi hat with 2GB of RAM and isn't anywhere near comparable to a full laptop motherboard. Try to make an adapter board that can handle all the IO necessary and with the low demand/production volumes of something that completely jerryrigged it'll only cost more if they want to break even.

    The only time low volume devices can compete on price with higher volume devices made from the same components is when they're made by hobbyists. Hobbyists who give out their time for free. Its completely and utterly selfish to just go around and demand everyone just give away their time for free rather than at least trying to cover their expenses. If you can't wrap your head around this concept then you should go away and realize people don't owe you their time and effort.
    They still have to convince people to buy it, This is essentially pissing away money instead of just getting the DC Roma I as now you are down a laptop. if they want to charge this much, it should at least be daily driveable in general​

    Originally posted by varikonniemi View Post

    It means nothing if it's slightly faster, if the cost is 3 times more and some of the features of pi5 come at even extra cost.
    Currently 2 issues plague riscv, Lack of vector support which is solved, and compiler optimizations. which will be on a per distro basis. The costs have already been affordable, and the riscv counterparts usually offer more features, better ram, gpu, hwdec etc.

    Leave a comment:


  • varikonniemi
    replied
    Originally posted by brucehoult View Post

    The 16 core version they've pre-announced for Q4 next year. It should be faster than Pi 5 / Rock 5 / Orange Pi 5, assuming the political situation is resolved.
    It means nothing if it's slightly faster, if the cost is 3 times more and some of the features of pi5 come at even extra cost.

    Leave a comment:

Working...
X