[Arm-netbook] Resource for quantifying the relative performance of the A20

[Matt Campbell]

Of course, there's no substitute for trying an A20-based board and
finding out if it's good enough for the tasks you need to do. And Luke's
recent videos suggest that it is indeed good enough for several common
tasks.

But I can't be the only one who's curious about how the A20 compares to
other processors, especially various x86 processors.

This is where UnixBench (https://github.com/kdlucas/byte-unixbench)
comes in handy. It's an open-source system benchmark suite for Unix,
going back a couple of decades. Someone took the time to run UnixBench
on a variety of computers, including the A20-based Cubieboard2, and
published the results here:

https://enchufado.com/proyectos/unixbench.html

Of course, that doesn't factor in graphics performance, but I thought it
might still be worth sharing.

Matt

_______________________________________________
arm-netbook mailing list arm-***@lists.phcomp.co.uk
http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook
Send large attachments to arm-***@files.phcomp

[Russell Hyer]

Well, I don't think the issue right now is whether or how it compares
with other chips (we have other chips already). Of course, if it's
libre competition, that's also for the good. I guess Intel peeps could
be redirected to https://puri.sm/

(You can check their website as they are also trying towards RYF certified, see:

https://puri.sm/road-to-fsf-ryf-endorsement-and-beyond/

)

And like academic arguments, multiple libre platforms is a good thing.
(Although it will still be kinda rare in real terms even if both
projects achieve it).

Russell
Not a real mathematician, just an applied geek

Post by Matt Campbell
Of course, there's no substitute for trying an A20-based board and
finding out if it's good enough for the tasks you need to do. And Luke's
recent videos suggest that it is indeed good enough for several common
tasks.
But I can't be the only one who's curious about how the A20 compares to
other processors, especially various x86 processors.
This is where UnixBench (https://github.com/kdlucas/byte-unixbench)
comes in handy. It's an open-source system benchmark suite for Unix,
going back a couple of decades. Someone took the time to run UnixBench
on a variety of computers, including the A20-based Cubieboard2, and
https://enchufado.com/proyectos/unixbench.html
Of course, that doesn't factor in graphics performance, but I thought it
might still be worth sharing.
Matt
_______________________________________________
http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook

_______________________________________________
arm-netbook mailing list arm-***@lists.phcomp.co.uk
http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook
Send large attachments to arm-***@files.phcom

[Luke Kenneth Casson Leighton]

---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68

Post by Russell Hyer
Well, I don't think the issue right now is whether or how it compares
with other chips (we have other chips already). Of course, if it's
libre competition, that's also for the good. I guess Intel peeps could
be redirected to https://puri.sm/
https://puri.sm/road-to-fsf-ryf-endorsement-and-beyond/

ohhh jaezus not again. *sigh* well at least they're trying with this:
https://puri.sm/posts/petition-for-intel-to-release-an-me-less-cpu-design/

what's pissing people off is that they're claiming *right now* to be
"libre" when in fact they are absolutely flat-out *NOT*. you can't
use the word "libre" - via the name "librem" and still require a
proprietary piece of firmware that runs on an NSA spyware-friendly
backdoor co-processor.

if they wanted people to trust them they should have put their foot
down and said NO, we will NOT EVEN SELL a product that has a backdoor
privacy-violating co-processor and attempt to label it "libre".

l.

_______________________________________________
arm-netbook mailing list arm-***@lists.phcomp.co.uk
http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook
Send l