On Saturday, 14 September 2013 00:36:05 UTC+8, James Reeves wrote:
> On 13 September 2013 08:54, Mikera <[email protected]
> <javascript:>>wrote:
>
>> Either way, if Clojure's semantics prove to be a fundamental issue for
>> performance, then I think it is better to start work to improve Clojure's
>> semantics (perhaps targeting 2.0 if we think it's a really big breaking
>> change). This would be better, IMHO, than forever accepting semantics that
>> prevent idiomatic code from ever being truly fast. I'd rather see a bit of
>> breakage and fix my code when upgrading to Clojure 2.0 than be stuck with
>> poor performance forever.
>>
>
> Out of curiosity, what is the performance hit in these cases?
>
Here's a quick highly artificial benchmark I did with Caliper in Java
(Obviously can't benchmark in Clojure since we don't have proper float
support....)
0% Scenario{vm=java, trial=0, benchmark=Float} 267222.79 ns; σ=1629.43 ns
@ 3 trials
50% Scenario{vm=java, trial=0, benchmark=Double} 416110.28 ns; σ=2087.75 ns
@ 3 trials
benchmark us linear runtime
Float 267 ===================
Double 416 ==============================
vm: java
trial: 0
Obviously this is just a microbenchmark, but it fits my general experience
that floats are a reasonable bit faster than doubles, typically 20-100%
(doubles are closer when it is pure number crunching since 64-bit CPUs are
actually pretty good at doubles, floats have a bigger advantage when you
are manipulating a lot of data points and hence memory bandwidth matters
more)
Code here for those interested:
src/test/java/mikera/vectorz/performance/FloatVsDoubleBenchmark.java
>
> Floats obviously save memory, and I believe they're also between 15% to
> 40% more efficient for division, at least on Intel CPUs. Is there also a
> cost to be paid passing doubles to OpenGL? Do you know what that cost is?
>
>
I'm not exactly an expert in OpenGL, but I would make the following points:
- Many GPU applications are actually memory-bandwidth bound (the compute
power is insane....). So using floats means you can pump through 2x as much
data to keep your GPUs busy
- Hardware and driver support is usually not on a par with floats, even if
it is technically supported by the OpenGL API
This is an interesting article to back up the latter point: For example on
NVidia GeForce hardware doubles are 1/8 the speed of floats
http://blog.hvidtfeldts.net/index.php/2012/07/double-precision-in-opengl-and-webgl/
--
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to [email protected]
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
---
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
