[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CALvmYyA6n8BTt5WR8B3ApguW1cO08mAZVWmC8hrzpDDunm+esw@mail.gmail.com>
Date: Mon, 25 Mar 2019 13:56:22 -0400
From: philip holbrook <fhlipzero@...il.com>
To: john-users@...ts.openwall.com
Subject: Re: Jetson nano - Like the Xeon PHI?
yea not worth it for sure, simple math.
That has 125 cuda cores, a couple year old 970 has 1600, i love the single
board computers and all but its just not powerful enough
On Mon, Mar 25, 2019 at 1:44 PM Solar Designer <solar@...nwall.com> wrote:
> On Mon, Mar 25, 2019 at 10:58:55AM -0400, Rich Rumble wrote:
> > Wondering if a cheap device like this could be used to create a nice
> > cracking rig on the cheap?
> >
> https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-nano/
>
> No. It will be either slow (few boards) or expensive (many boards).
>
> I am puzzled by the "like Xeon Phi" in the Subject since this device is
> dissimilar to Xeon Phi... except in that neither is cost-effective for
> our use case compared to typical large(r) gaming GPUs. I doubt that's
> the similarity you had in mind.
>
> Per my rough estimate, it'd take 30 or probably more of these boards to
> match one high-end GPU. That would cost $3k, but you could have had a
> high-end gaming GPU for $1k. It's similar with Xeon Phi - a ~$3k device
> that is a lot slower than a $1k gaming GPU for most of our needs.
>
> There may be a small percentage of hash types where a cluster like this
> would perform faster, especially through use of the CPUs - e.g., perhaps
> it'd have an advantage at bcrypt - but then a cluster of Raspberry Pi 3
> will perform similarly at a fraction of the cost.
>
> Alexander
>
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
