Rui Nuno Capela wrote: > Ralf Mardorf wrote: > >> Hi Rui :) >> >> thank you. Since I retrieved and wrote emails 2 weeks ago, I made music >> using Qtractor. >> >> > > that's great! awesome. do you mind sharing the pieces? :) >
Yes, I can upload it and share the link, but I prefer to do that, when I'm fine with the mixing and the song is a finished song. >> The external MIDI and audio equipment is well cultivated partway >> self-made home recording equipment (as I often wrote, I was working for >> a well known manufacture for studio equipment at development and >> manufacturing), no problem is caused by this. Using the Atari ST and >> analogue tape, there is no jitter and no sync problem, and no means no. >> I still didn't test the mobo so hard whit Windows. I only installed >> Windows because of the rt-audio troubles I've got when running Linux. >> >> At the moment I still have to join to older reports I made. >> >> 1. I needed to increase jackd's 'p' option to 512, to be able to go on >> working. For the mastering I'll increase 'p' again, because there are >> still xrun like sound effects. >> 2. An easy and modest test song that still isn't finished, nearly needs >> 100% CPU, while the sound quality of the mixing still isn't >> professional. A heavy mixed Windows test-song needs less than 80% CPU >> and the mixing (nearly) has professional sound quality. >> 3. I recorded electric guitar and external MIDI equipment at lower 'p' >> than 512, but I needed to manually fix the timing by editing note for >> note, something that for most recorded material isn't possible. >> >> Resume for the moment: >> >> Usage of Qtractor with DSSI and LADSPA cause no jitter and no problems >> with sync, but until now I wasn't able to finish and mix a song, there >> is trouble because of the CPU load and still the mixing quality is bad. >> >> > > where and how you're reading this cpu load? note that the percentage read > on the qjackctl panel is *not* system cpu load, it is jackd own process > dsp load and you can only compare that within same system > software/hardware setup. iow, there's no such figure anywhere else for you > to compare with. > I didn't run QJackCtl, I got the information by JAMin and top and especially because I heard dropouts (without and with getting xruns) or LADSPAs won't run any more etc. ... I guess to check my system it's good to compare with a Windows install. I can say that my hardware isn't broken, because I can do a song on Windows. I should do the same song for Linux and Windows. > also, you should not equate sound quality to cpu% usage. just to give you > a hint, i am aware that qtractor mixer widget sucks, and sucks hard the > more tracks you have in the show--just try to hide it while playback is > rolling and look at the tremendous cpu% usage drop :P > > however, this cpu consumption is purely in the gui--it doesn't, or > shouldn't, affect whatsoever the dsp audio processing, which is supposed > to be running in real-time scheduling priority (cf. jackd -R). > I have to add, that my board wasn't fine with Linux, since I bought it a year ago. i compiled and patched several kernels and applications myself but always run jackd. Now I run jackdmp. With jackd I got time-outs and clients were disconnected all the time. Maybe there still is an incompatibility for my mobo and Linux rt-audio, even if I don't get any time-outs and clients are never disconnected any more. > the quality issue bothers me though. i really wish to narrow where that > quality drop is due. is it qtractor's fault or is it something else? it > would help if you could put that up to something less subjective or not so > vague. > I should master the unfinished song. It's vague because I don't know the reason. I've some speculations, e.g. when making music on Linux I'm using a lot of free soundfonts and sometimes they remind me of my Roland MT-32. Stuff like the Roland MT-32 (I guess other people e.g. are using a Yamaha FB-01) are a good addition to expensive equipment, but it's no fun to use such bad sound sources without professional sound sources too. It might be that the reverbs or EQs do bad phases or ... or ... I'll try to master what I've got until now, maybe today, maybe at the weekend and I'll hand the link in later. >> Recording external equipment not only MIDI, but also guitar, vocals etc. >> isn't possible because of bad sync and jitter. >> >> > > not possible? now that seems a little bit hard to believe :) i do that all > the time and besides the fact that qtractor does no latency/delay > compensation, it surely is possible to record from external gear. > I can't compensate the delay comfortable and especially for external MIDI equipment it's impossible because of extreme jitter. I get dropouts when jackdmp does 512 frames/period at 2 periodes/buffer on soft or "normal" mixing, while ASIO is fine with 128 samples/buffer on hard mixing. This is vague, I know ;). >> Just to check if my computer hardware is fine, I compare Linux with >> Windows! (I'm not dissing Linux and I never came from Windows) >> >> On Windows latency on heavey usage of VST and VSTi is lower than it is >> for DSSI and LADSPA on Linux for minimal usage. >> On Windows CPU load at heavy usage of VST and VSTi is lower than it is >> for DSSI and LADSPA on Linux for minimal usage. >> On Windows the sound quality of the mixing by using VST and VSTi is >> (nearly) professional, but bad when using Linux with DSSI and LADSPA. >> >> > > i think you're comparing apples to oranges here. vst, dssi and ladspa are > just plugin interface specifications. the quality you refer seems to be > greatly subjective and can only be blamed to the particular plugin > implementations in use. to my knowledge and opinion, there's way too few > examples of high quality ladspa and even fewer if any at all dssi plugins > out there. comparing that with the immense offer in the windows vst > universe is not being fair at all :) > This isn't a battle Windows vs Linux, it's just that it seems to be able to do a nearly professional CD quality recording with Windows, but not with Linux on my machine. Maybe I need to use bad LADSPA EQs, because of the CPU load. Maybe there are good EQs, but my CPU is overloaded when I try to use them. I don't know, but there were phasers, wahs and flangers that won't run. > as everything in software, you can have it cheap, fast and good, but you > can only have two of that simultaneously:) > I don't like the concept of Apple and Microsoft, I don't have money, but I can get Windows software for free, without using cracks. It has to be good and fast. I'll do nearly professional recordings, resp. if I visit a friend with better equipment the production from my computer, that has to be nearly professional, should be good enough, to make a professional master-tape in the studio of a friend. I have loss because of my envy24 sound card, microchip op-amps etc. but that can be compensated when I do the mastering on better equipment. At the moment the quality of the test-song seems to be too bad, to compensate this. There seems to be missing to much in the sound quality. There's missing liveliness in the sound, I can't describe it on English, I'm able to do it on German. >> It might be possible to get a good sound quality using Linux on my >> machine too, but it has to be said, I run Windows for days testing >> rt-audio productions and Linux for years testing rt-audio productions. I >> never heard a Linux production made by someone else, that was a mix of >> DSSI, LADSPA and external equipment, but all mixed in the box, that had >> a professional sound quality (links to music done in such a way are >> welcome and hints which LADSPAs are fine and which have a bad quality >> etc. are also welcome). I still will try to remix the unfinished >> test-song and try to get a good sound quality. >> >> I don't think I'm the only one having trouble with sound quality, >> someone from the 64 Studio list made a mixing that is very well done, >> but anyhow it sounds like a 4-track tape recording and not professional. >> He decided to master his recordings at a professional mastering studio. >> My involuntary choice would be Windows :(, instead of a professional >> mastering studio. >> >> I don't know if Windows is fine with sync and jitter for external >> equipment or if it will be the same like it is for Linux, but even if it >> should be like it's for Linux, it's still possible to keep the latency >> lower with Windows and I guess a lower latency on Windows will have the >> same positive effect on jitter and sync, it has got for Linux. >> >> >>>> By the way, here is the output of rcirq: >>>> >>>> http://lists.64studio.com/pipermail/64studio-users/2009-June/003054.html >>>> >>>> >>> if your primary audio interface is the "usb" one, try to put it ahead of >>> the "snd" in the priority sequence--see your rtirq.conf, like this >>> >>> # IRQ thread service names >>> # (space separated list, from higher to lower priority). >>> RTIRQ_NAME_LIST="rtc snd usb i8042" >>> >>> change that to: >>> >>> RTIRQ_NAME_LIST="rtc usb snd i8042" >>> >>> and restart. >>> >>> byee >>> >> No, my primary sound device is an Envy24 based PCI card. The USB device >> is for MIDI only. I guess once upon a time I got better sync and/ or >> less jitter when not using RTC or what clock ever, but sync to PCM >> output or what it is called for Rosegarden's sequencer clock. >> >> > > ok, try using "usb3" instead of just "usb" in that rtirq.conf line. that > will make your particular usb midi device ahead of the usb crowd. just > remember that will probably change depending on which usb host port the > device is plugged. > Okay, I'll mark this Email and do this during the week. I also will try to get the f...ing MIDI cable for my envy24 card soldered correctly. The game port MIDI cable I soldered is fine with game ports, but unfortunately not with the port of my TerraTec card. Maybe they don't give ground to all pins, were ground should be. I couldn't write to TerraTec, because they are blocking emails from my provider (my provider is on pool position in the spam chart show). The USB-MIDI is also tricky with Windows. Sometimes I need to restart the sequencer for Windows, to get the USB-MIDI input in sync again. I don't need to do this for Linux. But on Windows this is the only trouble I know until now. > getting back to midi timing. rtc was indeed problematic before, moreover > on a rt-preemptive kernel (which still is highly experimental atm). to > make things worse the vanilla system timer is bad as easy, mostly due to > lousy default resolution (250hz). for quite some time, having the midi > timer dependable on the pcm device timing was the reasonable choice but > then it made jitter a function of the audio period/buffer size. not good > and open for too many ymmv issues. > I guess pcm sync was better on Rosegarden, than RTC sync. > that's why one could only go with 1000hz kernel timers. newer kernels (alsa) > have this default high-resolution timers option which might be a blessing. > I had jitter and sync problems with 1000Hz kernel timers. It was default for the 64 studio 2.1 one and I also set it for self compiled and patched kernels. > you see, linux audio/midi is and has been a moving target, with too many > ups and downs, and still a swamp of good and bad software solutions and > hardware combinations. > > in many ways, it is far, too far yet, from a complete or finished > business. i wonder if it will ever be ;) > > take care that i do this linux audio/midi stuff as my hobby, but seriously > enough. all your assessments are and will be greatly appreciated, ntl, as > long as you have some patience and time to spare, of course. it works > exactly the same to me. > > i've ditched windows 10years ago when linux was a obviously a joke wrt. > audio or midi, music production. if you have something at least workable > these days that's hugely thanks to a very few stubborn heads out there. > and believe me, they're quite a few worldwide, numbered by fingers of your > hands (i've just been following the movement close enough in these last > years so say just that ;) > > cheers > Fair or not fair, I compare the quality of different operating systems from the view point of my needs. Linux don't need to become perfect (no other OS seems to be perfect too). There are some very good new applications for Linux, that make me hope that Linux can fit to my needs, unfortunately they are buggy or uncomfortable. Jconv, some mixing consoles, guitar multiFX today seems to be good, but seldom they are part of package repositories. VST support by default would be a good thing ;), if VSTs are really fine on Linux. Okay, I'm babbling now, so I better stop the mail here. TODO (before I try to change something for MIDI): I'll master and upload what I call bad sound quality, that might be caused by some technical reasons and not because I did it bad. Please be patient too. I don't like to publish test songs, especially not unfinished stupid once :S, but I'll do it for once. Thanx, Ralf PS: Hm?! Offlist?! I removed the 64 Studio developers list, but I guess it might be something still for the users list. By the way, all bad starts with this (listen to the link of my actual signature): -- http://www.dailywav.com/1002/beginning.wav _______________________________________________ 64studio-users mailing list [email protected] http://lists.64studio.com/mailman/listinfo/64studio-users
