lekt player

Anything to do with computer audio, hardware, software etc.
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: lekt player

Post by jesuscheung »

actually i love that bass of sxb!

loving these
3.9 small
3.0 masm
3.92 sxb

still haven't tried other versions
lekt
Posts: 1102
Joined: Thu Oct 10, 2013 4:52 am

Re: lekt player

Post by lekt »

jesuscheung wrote:lekt have you tried on your laptop
1. lock desktop
2. power plan -> close lid -> does nothing
3. close the lid of laptop

all things better
yes, SQ improved, very good method for listening.

i see v3.9 256 small have special sound stage, i called it "small" coz stage size is smaller than others. but its size is correct. many other versions still a bit 3D-effect, very difficult remove this effect.

JC, i tried compare binary content of files builded by the same code, different only in /COFF File Header/TimeDateStamp (4-bytes, dword). documents says: TimeDateStamp: The low 32 bits of the number of seconds since 00:00 January 1, 1970 (a C run-time time_t value), that indicates when the file was created.
what OS loader does with this number when loading file into RAM? difference files (from the same code) have difference sound. tried edit PE's TimeDateStamp = 0, seems sound better. maybe there's most true sound created by this code? set TimeDateStamp = TimeDateStamp of other version, sound have colour like as refference version.
try refference your favourite versions, interesting test.
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: lekt player

Post by jesuscheung »

lekt wrote: i see v3.9 256 small have special sound stage, i called it "small" coz stage size is smaller than others. but its size is correct. many other versions still a bit 3D-effect, very difficult remove this effect.
....
tried edit PE's TimeDateStamp = 0, seems sound better. maybe there's most true sound created by this code?....
shall MOD it!

----
3.9 small has less bass. maybe why soundstage more easily coming together?

in the old days i modded c7 with PE, it removed some bass. stage got "smaller" also.
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: lekt player

Post by jesuscheung »

lekt wrote:....tried edit PE's TimeDateStamp = 0, seems sound better. maybe there's most true sound created by this code? set TimeDateStamp = TimeDateStamp of other version, sound have colour like as refference version.
try refference your favourite versions, interesting test.
some earaches disappear here! maybe apply this setting for all new versions?
lekt
Posts: 1102
Joined: Thu Oct 10, 2013 4:52 am

Re: lekt player

Post by lekt »

think TimeDateStamp = 0 is best, seems internal hash procedure reject this null number so then sound better.

uploaded:
v3.11 256 sse
v3.11 256 mlp
v3.11 256 4lp

v3.11.1 256 4lp mov
v3.11.1 256 4lp xor

v3.11 256 sse have good balance, maybe it is best?
i like sound stage of v3.11 256 mlp, soft and thin sound.
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: lekt player

Post by jesuscheung »

lekt wrote:think TimeDateStamp = 0 is best, seems internal hash procedure reject this null number so then sound better.

uploaded:
v3.11 256 sse
v3.11 256 mlp
v3.11 256 4lp

v3.11.1 256 4lp mov
v3.11.1 256 4lp xor

v3.11 256 sse have good balance, maybe it is best?
i like sound stage of v3.11 256 mlp, soft and thin sound.
i like 3.11 mlp.

-------
TimeDateStamp = 0 all things better here, but decay lesser a little

but if TimeDateStamp != 0, has some earaches like 3D
maybe the extra decay stetches the sound

feels that TimeDateStamp = 0 is correct
lekt
Posts: 1102
Joined: Thu Oct 10, 2013 4:52 am

Re: lekt player

Post by lekt »

jesuscheung wrote:...
TimeDateStamp = 0 all things better here, but decay lesser a little...
yes, sometimes TimeDateStamp makes sound like as better (ie. v3.9.1 256 sound is good, decay), but need accurate test, think TimeDateStamp creates sound effect if !=0.
i don't know what OS loader does with this TimeDateStamp, maybe hash...
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: lekt player

Post by jesuscheung »

quickly tried 3.1x to 3.8x

keeping lekt.exe v3.1 256 raw
love the simple analog sound. nothing overdone- power, decay, hard/soft, thin/rich, depth etc
lekt
Posts: 1102
Joined: Thu Oct 10, 2013 4:52 am

Re: lekt player

Post by lekt »

jesuscheung wrote:quickly tried 3.1x to 3.8x

keeping lekt.exe v3.1 256 raw
love the simple analog sound. nothing overdone- power, decay, hard/soft, thin/rich, depth etc
v3.1 uses almost variables in memory, lesser register usage. maybe not good speed for data copying but ignore (or lesser) register dependencies, sound more natural.
now need balance/optimize beetwen memory usage and register usage. Maximize register usage not always good coz too long time holding register then CPU replaces with other free resource (other registers, memory) for doing this in background, we don't see this work like as writen in code. register dependencies affect parallel executing instructions.
Register usage is art, need find some principles and learn accurately.
jesuscheung
Posts: 2491
Joined: Mon Oct 07, 2013 11:09 pm

Re: lekt player

Post by jesuscheung »

lekt, can't wait to hear your registry code!

love 3.0, 3.1 raw, 3.9 small, 3.9.2 sxb, 3.11 mlp
mlp probably is the best one
Post Reply