Beiträge von tonstudio96

    The piano's hardware is pretty complex and cannot be modelled with common synth's architectures. Think only of the interaction of the strings within a triple string group especially when played in sostenuto mode. You will have to apply energy interchange effects. I did an approach using specialized equations in own DSP System taking such effects into account. The problem is the processing power and the money such a system would cost.


    http://www.96khz.org/oldpages/fpgapiano2.htm

    Wow, today one needs an attorney and special FBI agents among the personal friends just in order to like a musical instrument safely.

    Seriously, what do you expect from facebook? This is no private university network anymore as it used to be, but a company with economic interests. Their shares are out and trade at stock already, aren't they? I see no need to inform a private US company about my likes and my business, like many "modern" people obviously do, when they show interest for new goods, material and even start negotiations about business activity and freelancer projects as I have seen recently. If they want to make money by trading based on such information, they have to search for it themselves.

    it would not sound nearly as good as the TI sounds..


    Hm, if the software stays the same, it should be right. Basically it is only a matter of calculation power, whereby it might be an enourmous work to change a DSP software to optimized Intel or MAC code. I recently did that for a radar app moving the firmware of formerly used 8 ADSPs to on INTEL core running on a VPX. Works perfect.


    Theoretically you could run every music software on a large PC and with current 64Bit OS they should have also enough speed for that, but I doubt if it makes sence to do so. CPUs for PCs and also their hardware (especially RAM) is not really optimized for signal processing and most probably will never be. PC based apps are usually low cost versions in comparison to dedicated DSP hardware. Also even with 64Bit MCUs current PC music platforms seem to be busy already processing the host app :)


    If synthesizers had more demand to operate in parallel, GPU usage was another idea, but as I found with another recent project, the GPU also becomes quickly expensive too or you have to add an extra device of it, if you want to have relevant power.


    I could imagine running more SynthSoftware on anonymous DSP hardware inside PCs, since PCI offers much better integration and a higher data speed interface than accessing outboard. On the other hand, USB 3.0 has all to quickly interlink any hadware to your music host PC.
    I guess a soft synth won't be much cheaper anyway (as long as you pay for it and do not use cracked sw), and if you add a good midi controller and ADC/DAC hardware and the 30% price of the PC the softsynth occupies, you almost have the price of a hw synth.

    This sounds like a problem of arrangement, in detail the balance of the sound of all your instruments in the band. Many people can't hear themselves neither in recording situations nor live situations and even not on the CD, or they have the impression, that they are not present enough - so they tend to increase their loudness, possibly by compression, even increasing the problem:
    The higher the average level (= lower dynamic) the more is the level of importance and slight changes will let one instrument dominate and the another one "die". Try to use uncrompressed music and use in-ear-monitoring with increased individual volume. When rehearsing identify those passages of a mix, where individual instruments were to loud or not loud enough and change this individually. The musicians have to learn to do this. A Trick: Try to rehearse with the very lowest level possible and make sure still all instruments are present they way they have to for the particular role and point of time.


    For Keyboarders and Synthieplayers fo example, one has to distinguish the parts in the mix where accompaingment and supporting is required and when solo or emphasis is done. Usually this is not possible to do without some loadness change during songs, and this often requires touching the volume knobs for certain channels (left hand, right hand, bass etc). Sometime a foot controller helps.
    General emphasis by compression, presence-EQ-ing or enhancing by exciters are of no advantage and will lead to the loudness battle.
    In some cases, expanders are usefull to emphasize an instrument's activity

    Similar with me: Returning user, allready having participated in the old forum / music mailing list. Having used Waldorf and Virus Synths from the mid of the 90tees - still owning a "B" and "Q.". Running a studio for production and recording (mostly acoustical musics). Primary business is electrical engineering / digital signal processing - mostly imaging, rarely audio.


    Created own sounds and music / sound algorithms, starting in the 80tees with C64, since the 90tees in C at Microprocessors (TMS320X, DSP5630X), since 2003 with FPGAs. Created an own music workstation with virtual audio processing DSPs, midi like tone generator and sound synthesis with currently up to 1024 voices polyphony at 192kHz (4096 at 48kHz) spreaded over 6 FPGAs.

    Why do people prefer knobs all over their synths?


    Live Performance? Ever did that?
    And also with Computer-only music, you will need some physical controllers anyway, no matter if you add a USB-based device to your computer and route it to VSTs, Rack-Modules or whereever - or have MIDI-Controllers in between the source and destination.
    I do not excpect you to use just a mouse?


    >the manual for TI is like a...
    Really that bad?

    You sold it or gave it back because of midi routing issues?


    Well talking for Virus B, it responds to 16 tracks so, making it a multitimbral player is just a matter of midi wiring logistics.


    I am using a hand made midi router in an electronic device to seperate a track into 16 channels to use e.g. 16 drum tracks simultaneously coming fomr channel 10 in MIDI tracks. There are a number of instrument hardware units out doing this too, if somebody cannot do it from software.


    Without drum tracks, it should be even easier to use 15 tracks simultaneously apart from the fact of course, that the synth has limited number of voices.


    This should work with all synths and all midi setups, so I consider issues in software handling. Either your tracks are not joined correctly to one physical midi channel or the track parameters intersect with each other.

    ROM banks are most useful for patches which would most probably never change, such as sounds you want to use again and gain. An example was to create patches covering soundbank constellations know outside the virus world, where instrument numbers refer to standards like GM. For exampe you could reserve one bank for GM and another for GM drums, so the virus can play standard midi for rehearse.

    Ok, thanx for the information. For the moment the Access Virus Version seems to crash for some reasons, while the Emagic Version works. It also connects to the Virus and I can edit patches. Both are namend 3.0. (Virus Setup 01.03.2001 9:47)


    In the Internet I found 3.04 on some mirror pages. But they are using download functions which seem to be a bit unsafe to me.


    What looks fine is this here: http://www.solar3d-software.com - but it is for the virus only.

    BTW this is a great example why one should keep an old Pentium PC sitting around with DOS, Windows 3.1, Windows 95/98 and/or Windows 2000.

    Absolutely, I got 2 of em behind me :) One is my old music PC wich now has a new drive, a fresh Win98 and double RAM - but: No Sound Diver anymore.


    Well I did not use it for years now, to be honest, but need it now again. I remember I got my Virus prior the SD-Access period, since I did not get a CD at that point of time. I ordered one from Access for around 40,- AFAIR and used it several times - but it is gone. I think it was called CD Virus 4.0 or similar and was a kind of enhanced version of the Emagic Version 3, is that correct?


    Is there a chance to reveal this? I need a Version which operates with either Win98 or XP. On my old CD-Roms, I found a SD Version from Emagic, I never used. It came along with a device, I think. It is labeled to go with W95 only but obviously works in Win XP fine. I could register the Virus B and also some of my sound modules.


    So think, the Access Version might work on XP too?


    -------------------


    update: Found the CD (with that interesting yellow :) , nstalled it on XP and started scan: Most Patches were found except including Bank A/B - then it crashed. Seems a known Problem:
    http://virus.info/forum/index.php?page=Thread&threadID=1139


    Anybody has experiences in running that piece of software in a Virtual Environment on a modern host?

    - Multiband Compression
    - Harmonic Equalization
    - Volume maximization / brickwall limiting
    - "Polishing" to create brilliance

    Some of them are channel issues, and cannot be done in the master section. Anyway you can use the plugin though. I had recommended IZOTOPE to you, too :) Well, what is easy to use: Waves Native Gold - L-Compressor + C4/C6 before. For a all in one tool, try T-Racks. It pretends to use tube emulation - I used that in former times effectively. (Tube only for the voice channel , never on the full material).


    What you should find and use in the plugin for final mastering is soft only MBC, dynamic increase by soft limiting only, and invisible loudness correction with a very slow compressor before the whole mastering section, and last but not least: dithering / noise shaping for 16 bit - assuming you produce for CDs.

    After browsing the Access Site a bit more carefully, than the last time, I found, that obviously V 4.9 from the year 2008 is the latest and last OS for the Virus B. - Surprisingly I found that my Virus comes up with 4.5! allthough I thought, I had done an update in year 2009. I could bet against the devil, I did that.


    Anyway, just to exclude any misunderstandings, I'd like to ask, how an OS4.9 virus B opens at startup? Is there the OS number 4.9 explicitely named? - The point is the following: When browsing the OS4.9 archive, I found also hints to 4.5, namely the manual, patches and a 1024 patch mid. Also the update is named, 4.9050 and this might be read as mayor/minor 4.50. I definetely want to avoid an unnecessary update, since I read about some stopps and problems.


    So can anybody confirm, that 4.9 will have to appear?


    My System shows (double checked) only the figures : " OS 4.5 S " and nothing else. - I was absolutely sure, I had the final update allready. (?) For example, I have the 1024 patches on board - maybe I overlooked that or the update did not work, or what so ever.


    -----------------------------------------------------------------------


    Regarding the 1024 patches: I assume, they did not change and do stem from the 4.5? - Assuming, I will do an update to 4.9 - is it necessary to perform also the 4.5-1024patches update? Maybe this is an extra flash area and need not to be written, because already done by the prior update.(?)


    -----------------------------------------------------------------------


    Regarding the RAM-patch this does not bother me, since it does not belong to the flash / OS update.


    -----------------------------------------------------------------------


    Regarding Player: As I learned from here:
    http://virus.info/forum/index.php?page=Thread&threadID=1859
    http://virus.info/forum/index.php?page=Thread&threadID=3383


    It is save and most likely to work using MediaPlayer as MIDIfile-Player with 20 - 40 bpm.(?)


    -----------------------------------------------------------------------


    Anything, I forgot? - Should the Virus warm up for some minutes before starting?

    I had also said before, that a HW-compressor outside is the best solution, but in the meanwhile I found certain situations, were I liked to compress (in terms "auto level adjust") a channel BEFORE it affects others / is affected by other channels, especially before it is mixed together with other audio channels. Once it is routed to output and premixed, the chance is away.


    In my DIY Workstation, I have both: MIDI-Compressor preSynthesis and Audio-Compressor postSynthesis, as well as individual EQs. Also the Return, especially the effect returns do have compression options. This helps much.


    One example for compressors to go with very variant signal levels, caused be ring modulation and cross matrix channel effects and such tinhgs:


    I typically use a setup of two chained compressors, namely the first with very slow settings performing an AGC, the second is a very fast one performing soft limiting. In this way, I smoothly can continously autoadjust 3dB-6dB in avarage level and another 6dB-9dB peek without getting to strong sound distorsion - leading to two aspects: The fast Compressor operates mostly in a simular area, also with variant signals and the the sound become less mid time variant if I change filters or add channels.


    Especially adding channels and do modulation might easily overcome the dynamic and required real time adjustment of the patch valume to make the track still meet it's environment.