- About Scala
- Documentation
- Code Examples
- Software
- Scala Developers
halfway to nothing
Sat, 2011-05-21, 20:01
Have now shaved more than half the startup time relative to 2.9.0. Here we are oscillating between it and r25007.
% while true; do time scala29 -J-d32 -nc -e '()' ; time scala3 -J-d32 -nc -e '()' ; done |& grep real
real 0m3.322s
real 0m1.565s
real 0m3.166s
real 0m1.617s
real 0m3.168s
real 0m1.566s
real 0m3.167s
real 0m1.617s
real 0m3.163s
real 0m1.567s
real 0m3.268s
real 0m1.615s
real 0m3.164s
real 0m1.620s
real 0m3.164s
real 0m1.574s
real 0m3.217s
real 0m1.612s
real 0m3.271s
real 0m1.621s
Sat, 2011-05-21, 20:17
#2
Re: halfway to nothing
On Sat, May 21, 2011 at 8:01 PM, Paul Phillips wrote:
> Have now shaved more than half the startup time relative to 2.9.0. Here we are oscillating between it and r25007.
Nice! Will this be part of the next bug-fix update or will it wait for
the subsequent one?
Best,
Ismael
Sat, 2011-05-21, 20:27
#3
Re: Re: halfway to nothing
Woot, nice work Paul!
On May 21, 2011 9:05 PM, "Paul Phillips" <paulp@improving.org> wrote:> Oh, it's even more pronounced if I don't flip to 32bit. And here I thought I was rigging it in my favor.
>
> % while true; do time scala29 -nc -e '()' ; time scala3 -nc -e '()' ; done |& grep real
> real 0m4.537s
> real 0m1.854s
> real 0m4.328s
> real 0m1.853s
> real 0m4.220s
> real 0m1.852s
> real 0m4.255s
> real 0m1.904s
> real 0m4.421s
> real 0m1.954s
Sat, 2011-05-21, 20:37
#4
Re: halfway to nothing
On 5/21/11 12:08 PM, Ismael Juma wrote:
> Nice! Will this be part of the next bug-fix update or will it wait for
> the subsequent one?
This will wait.
Sat, 2011-05-21, 20:47
#5
Re: halfway to nothing
On Saturday May 21 2011, Paul Phillips wrote:
> On 5/21/11 12:08 PM, Ismael Juma wrote:
> > Nice! Will this be part of the next bug-fix update or will it wait
> > for the subsequent one?
>
> This will wait.
Always holdin' out on us!
RRS
Sat, 2011-05-21, 23:37
#6
Re: halfway to nothing
3.0?
On Sat, May 21, 2011 at 3:13 PM, Paul Phillips <paulp@improving.org> wrote:
On Sat, May 21, 2011 at 3:13 PM, Paul Phillips <paulp@improving.org> wrote:
On 5/21/11 12:08 PM, Ismael Juma wrote:
> Nice! Will this be part of the next bug-fix update or will it wait for
> the subsequent one?
This will wait.
Sun, 2011-05-22, 12:17
#7
Re: halfway to nothing
On Sun, May 22, 2011 at 12:29 AM, Josh Suereth <joshua.suereth@gmail.com> wrote:
3.0?Hopefully not that long... -- Martin
On Sat, May 21, 2011 at 3:13 PM, Paul Phillips <paulp@improving.org> wrote:On 5/21/11 12:08 PM, Ismael Juma wrote:
> Nice! Will this be part of the next bug-fix update or will it wait for
> the subsequent one?
This will wait.
--
----------------------------------------------
Martin Odersky
Prof., EPFL and CEO, Typesafe
PSED, 1015 Lausanne, Switzerland
Tel. EPFL: +41 21 693 6863
Tel. Typesafe: +41 21 691 4967
Sun, 2011-05-22, 13:57
#8
Re: halfway to nothing
Ah, I was refering to the script name.
On May 22, 2011 7:14 AM, "martin odersky" <martin.odersky@epfl.ch> wrote:> On Sun, May 22, 2011 at 12:29 AM, Josh Suereth <joshua.suereth@gmail.com>wrote:
>
>> 3.0?
>>
>> Hopefully not that long... -- Martin
>
>
>>
>> On Sat, May 21, 2011 at 3:13 PM, Paul Phillips <paulp@improving.org>wrote:
>>
>>> On 5/21/11 12:08 PM, Ismael Juma wrote:
>>> > Nice! Will this be part of the next bug-fix update or will it wait for
>>> > the subsequent one?
>>>
>>> This will wait.
>>>
>>
>>
>
>
> --
> ----------------------------------------------
> Martin Odersky
> Prof., EPFL <http://www.epfl.ch> and CEO, Typesafe <http://www.typesafe.com>
> PSED, 1015 Lausanne, Switzerland
> Tel. EPFL: +41 21 693 6863
> Tel. Typesafe: +41 21 691 4967
Sun, 2011-05-22, 16:47
#9
Re: halfway to nothing
On 5/22/11 5:56 AM, Josh Suereth wrote:
> Ah, I was refering to the script name.
Oh, I didn't realize I'd exposed that. It doesn't mean anything. I
have a zillion aliases for scala versions, and scala29 was my latest
tested build of 29 until 2.9.0 came out. Now I need a new alias for my
latest build, and scala210 is annoying in terms of finger mechanics,
file completion, lexical ordering, etc. whereas scala3 isn't.
Oh, it's even more pronounced if I don't flip to 32bit. And here I thought I was rigging it in my favor.
% while true; do time scala29 -nc -e '()' ; time scala3 -nc -e '()' ; done |& grep real
real 0m4.537s
real 0m1.854s
real 0m4.328s
real 0m1.853s
real 0m4.220s
real 0m1.852s
real 0m4.255s
real 0m1.904s
real 0m4.421s
real 0m1.954s