Re: Windows performance

[Trevor Daniels]

> Phil Holmes wrote Thursday, April 16, 2015 3:43 PM
>>
>> From: "Trevor Daniels" <address@hidden>
>> Sent: Thursday, April 16, 2015 3:28 PM
>>>
>>> Phil Holmes wrote Thursday, April 16, 2015 2:00 PM
>>>
>>>> The performance of LilyPond 2.19.18 on Windows is _much_ better than
>>>> previous versions.  Some examples:
>>>>
>>>> A 26 page multi-score piece I've been working on:
>>>> 2.19.16: 114s to compile
>>>> 2.19.18: 52s
>>> ... [etc]
>>>
>>> That's remarkable.  I can see no change between 2.19.16
>>> and 2.19.18 that might account for this enormous change.
>>> The only effect of this magnitude which I've seen in the
>>> past is to do with setting up the font library when LP
>>> is run for the first time.  But that would be an increase.
>>>
>>> Any chance your 2.19.18 is using a different hard disk,
>>> an SSD maybe, which the others weren't?
>> 
>> No.  They're all installed on an SSD.  It's CPU limited anyway.
>> 
>> I thought 2.19.16 might be quicker because of the change to the compiler, 
>> but, as you say, have no idea why .18 is so much quicker than .16.  I'm 
>> pretty certain it's a genuine difference: I only noticed because the score I 
>> was working on suddenly appeared more quickly!
> 
> Well, I remain mystified, but I can confirm the speedup on my Windows Vista 
> laptop with a 4-page score:
> 
> With 2.19.16 this took 25.2  22.6  22.9 secs
> With 2.19.18 this took 12.6  12.5  11.8 secs
> 
> It is amazing, but very welcome!  In my career as a systems programmer, in 
> the days when performance was critical, I never saw an improvement of this 
> magnitude resulting from (presumably) a single change.  Usually we were 
> struggling to gain a percent or two.
> 
> Were there any changes to GUB between .16 and .18?

Further to this, I first established that the speed-up on Windows happened 
between 2.19.17 and 2.19.18.  One possibly relevant change between those two 
releases was the move from gs 8.70 to gs 9.15.  This increased the size of 
usr/bin from 49.2Mb to 58.1Mb, virtually all of the increase being in the three 
gs files, so this looks like a pretty significant upgrade.  Checking the 
GhostScript website for release 9 suggests it includes performance improvements 
and major changes to font handling.  This looks a possible candidate, installed 
courtesy of Masamichi HOSODA-san.  If so, yet more thanks and kudos to you, 
Masamichi! But wouldn't this show a speed-up on systems other than Windows?

Trevor