Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.os.linux > #76199

Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so)

Show all headers | View raw

Paul <nospam@needed.invalid> writes:

> On 9/12/2022 9:29 AM, 5GyYap52yQ1UGMWD wrote:
>> Dan Purgert <dan@djph.net> writes:
>> 
>>> Aragorn wrote:
>>>> On 09.09.2022 at 11:34, Dan Purgert scribbled:
>>>>
>>>>> CUDA is nVidia's GPU tech.  Used a lot in heavy math / supercomputing
>>>>> applications as I recall (and, of course, gaming).
>>>>
>>>> Nowadays, it's SPECIFICALLY being used for cryptocurrency mining.  And
>>>> Nvidia has now even begun developing GPUs which specialize in that.
>>>
>>> Yeah, but GPUs for crypto are so 2010.  Thought it all moved over to
>>> ASICs for that?
>> It  depends on  the crypto  that you're  mining. Bitcoin  has
>> definitely
>> moved into  ASICs but  Ethereum is,  or rather was,  sucking up  the GPU
>> market. I believe  they're moving to a different way  of doing things so
>> the demand for GPU for mining might die down soon.
>> 
>
> Ethereum is done with ASIC too.
>
> But I can't find any pictures to show you. And that's more than
> a bit weird, as I thought by now there would be a picture of an
> Ethereum ASIC board.
>

Odd. Maybe Ethereum did have ASIC  boards running on the network. Though
most of the people  that I knew who are doing  Ethereum mining are using
GPUs. Though I wouldn't blame them, ASICs is horribly expensive and it's
a long-term investment.  GPUs can  be repurposed for different tasks and
it's easier to unload them to the market.

> [...]
>
> The big mining operations may select GPUs, so they can alternate
> between mining Etherium and Monero. When Ethereum moves to Proof-of-Steak,
> you just re-load with Monero mining code, without even blinking.
> Or even Dogecoin, if there's a profit in it this week.
>

I  don't think  you  can mine  monero  using a  GPU.  RandomX is  highly
resistant to GPU mining.

> That to me, would be the main advantage of the GPU, is reprogramming.
> Nobody wants to throw away all their Bitmain boxes, on a whim.
> It takes a hell of a lot of mining, to pay for the box. Imagine
> having a $25 million hardware investment being wiped out instantly.
>

I agree.  But I  think ASICs  also contributed to  the stability  of the
Bitcoin hashrate over time. Having specialized hardware to do the number
crunching will  create dedicated  miners that  are "in  it for  the long
run."

There's pros and cons to that, of course, but I think the current steady
hashrate of Bitcoin had something to do with the miner market.

> Since Bitcoin is algorithmically stable, more boxes have been
> made for that as a result (Intel making a box, this late in the
> game, is an example). But it cannot stay that way forever.
> Technology marches onward.
>

This is what  you'd want on a  monetary good though. One  of the reasons
why I'm very averse with anything to  do with Ethereum is that they have
a track  record of being sporadic  with their design goals  and monetary
policies.

I mean, this  shift to proof-of-stake just shows this  trend. Though I'm
not that well  versed with smart contracts and if  this sporadic changes
in  design  has anything  to  do  with  that  since I'm  only  following
cryptocurrencies that aim to be digital cash (i.e: bitcoin and monero).

> What's impressive about the Intel chip, is the 0.355V core voltage
> to run it. A lot lower than my GPU at 0.9V core voltage. When I took
> engineering, nobody dreamed back then, that we'd get under a volt.
> It's getting ridiculous :-) This is one reason the power tracks
> feeding these silly things, are getting wider and wider. You
> could be talking hundreds of amperes, coming from the SMPS on
> the board.
>
> https://www.anandtech.com/show/17218/intels-next-gen-bitcoin-asic-called-bzm2-built-on-7nm-137-gigahashsec-at-25-w
>
>    Paul
>

That's cool.

What I find interesting, though, is  that intel is getting into the ASIC
game. I  wonder what  people will  think of that.  Maybe the  prices for
these things will  slowly drop as bigger  manufacturers produce hardware
for these applications.

-- 
Pointless meanderings in a bleak and lonely world.

Back to alt.os.linux | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

[crosspost] recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Soviet_Mario <SovietMario@CCCP.MIR> - 2022-09-07 14:13 +0200
  Re: [crosspost] recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Dan Purgert <dan@djph.net> - 2022-09-07 13:18 +0000
    Re: [crosspost] recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Soviet_Mario <SovietMario@CCCP.MIR> - 2022-09-08 00:39 +0200
      Re: [crosspost] recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Dan Purgert <dan@djph.net> - 2022-09-09 12:18 +0000
  Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) "J.O. Aho" <user@example.net> - 2022-09-07 16:51 +0200
    Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Soviet_Mario <SovietMario@CCCP.MIR> - 2022-09-08 00:42 +0200
      Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) "J.O. Aho" <user@example.net> - 2022-09-08 07:56 +0200
      Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Dan Purgert <dan@djph.net> - 2022-09-09 11:34 +0000
        Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Aragorn <telcontar@duck.com> - 2022-09-09 19:08 +0200
          Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Dan Purgert <dan@djph.net> - 2022-09-12 10:28 +0000
            Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) "J.O. Aho" <user@example.net> - 2022-09-12 14:36 +0200
            Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) 5GyYap52yQ1UGMWD <ehj46PkBWfBAng9C@VW28LtWn6wknpUMV.invalid> - 2022-09-12 21:29 +0800
              Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Paul <nospam@needed.invalid> - 2022-09-12 13:58 -0400
                Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) 5GyYap52yQ1UGMWD <ehj46PkBWfBAng9C@VW28LtWn6wknpUMV.invalid> - 2022-09-13 08:41 +0800
                Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Paul <nospam@needed.invalid> - 2022-09-13 03:20 -0400
                Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) 5GyYap52yQ1UGMWD <ehj46PkBWfBAng9C@VW28LtWn6wknpUMV.invalid> - 2022-09-13 16:38 +0800
            Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Aragorn <telcontar@duck.com> - 2022-09-12 20:58 +0200
              Re: recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Dan Purgert <dan@djph.net> - 2022-09-13 10:12 +0000
  Re: [crosspost] recommended slicers for 3D printer and 'filled' volume? estimation (drivers, consumes and so) Robert Heller <heller@deepsoft.com> - 2022-09-07 15:23 +0000
    Re: [crosspost] recommended slicers for 3D printer and 'filled' volume? estimation (drivers, consumes and so) Soviet_Mario <SovietMario@CCCP.MIR> - 2022-09-08 00:43 +0200
  Re: [crosspost] recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) not@telling.you.invalid (Computer Nerd Kev) - 2022-09-08 09:50 +1000
    Re: [crosspost] recommended slicers for 3D printer and 'filled' volume estimation (drivers, consumes and so) Soviet_Mario <SovietMario@CCCP.MIR> - 2022-09-08 03:18 +0200

csiph-web