I was doing some comparisions last night as I was working on a few projects, i have going. What i found is that sound fonts in gereral like different types of players.
some fonts like the sampler just fine, others like a true font player like SFZ or Jeskola, some like one other other better.
I found that strings like credenza strings like the jeskola player the best, if used in the sampler they were choppy and halting, if used in the SFz free player they sucked up tons of CPU, in the Jeskola player they were smooth and real sounding and very little CPU was used,
Some guitar fonts like sfz better than Jeskola, and played smoother there, in jeskola they seemed to arpegiate unnaturally, brass like borland better brass worked like the strings , while drums seemed to love the sampler just fine.
I sarted using the titanic GS font set it has an acoustic kit, a jazz kit a maple kit a power kit and a few more that are very great sounding even better than the s-kit i used for a while.
Any thought on this issue?
I e-maile the boss and he responded with the fact that the sample has not efects and some times is not as smooth on something, I thing that what it needs is a way to play larger layered fonts a bit better, and stream them smoother but it does not matter i have now learned
Strings and brass + Jeskola, Drums and bass and keyboards, + sampler, Guitars sometimes SFZ, other times sampler! all of this equals good sounds and easy CPU usage.
sound font reproduction
Hi there - this is a path I have also walked.. I use sfz and the MTS sampler and, like you, I find that they do react differently to soundfonts.
It seems that sfz has some tricks that MTS doesn't like - it is capable of reading two files at the same time (sometimes used to get a psuedo stereo effect). Equally, MTS is capable of operating with different size write and read buffers (so I was told) and sfz has trouble coping with this.
The bottom line for me is that sfz can produce some very nice sounds from soundfonts (but I find the sfz replay level is always rather low). However it does fall over (for instance, I can not play several sfz's at the same time that I record another sfz).
The MTS sample is more stable (I guess it should be!) but when you import and convert soundfonts, it does go wrong sometimes. I have seen it get the loop points wrong, or miss some of the layers. If you really want to use the MTS sampler with a particular soundfont, you can make your own patch and find and load each layer file. Where it gets to be a real pain is having to find and set the loop points for each sample. Hmmm - I tried that, no thank you
I have not come across Jeskola - perhaps I should give it a try.. do you have a web address for them? Oh, and is it free?
Regards
Pete H
It seems that sfz has some tricks that MTS doesn't like - it is capable of reading two files at the same time (sometimes used to get a psuedo stereo effect). Equally, MTS is capable of operating with different size write and read buffers (so I was told) and sfz has trouble coping with this.
The bottom line for me is that sfz can produce some very nice sounds from soundfonts (but I find the sfz replay level is always rather low). However it does fall over (for instance, I can not play several sfz's at the same time that I record another sfz).
The MTS sample is more stable (I guess it should be!) but when you import and convert soundfonts, it does go wrong sometimes. I have seen it get the loop points wrong, or miss some of the layers. If you really want to use the MTS sampler with a particular soundfont, you can make your own patch and find and load each layer file. Where it gets to be a real pain is having to find and set the loop points for each sample. Hmmm - I tried that, no thank you
I have not come across Jeskola - perhaps I should give it a try.. do you have a web address for them? Oh, and is it free?
Regards
Pete H
I found Jeskola at K-v-R it was $49 it will play 16 fonts like the SFZ= player, I only have the sfz free version, It does a good job mostly, but takes up far to much CPU. The jeskola seems to use much less power, but it does mess up some fonts, especially arpegiated fonts like guitar chords, or finger picking stuff. I am sure there is an adjustment but have not found it yet. The sampler is great for most fonts, it just seems to suck at strings, and brass, I never thought of the conversion process being the problem, and would have no idea how to fix that or even the will to try now, I intend some day when maony loosens up to buy a korg triton sound module, to take the fight out of midi and synths. but for now it is fonts, Man i gottta say that titanic GS set is awesome I can remembe where i downloaded it but it rocks for drums and stuff, it has some nice guitars and basses as well.