Kirk Hunter Studios FAQs, Tips and Peer2Peer Info Sharing Thread

[AKJ]

Hi,

so far, I found the following bugs in Diamond (yes, I have applied the update). Kirk, can you please test/verify and provide fixes:

1. VPRO Solo Brass: same bug as with Woodwinds (IntervaLive down does produce noise) where it was fixed. The VPRO instruments are important since some insturments are not present in the TVEC3 versions (i.e. trumpet a, trombone a and french horn a).
2. TVEC3 Woodwinds: English Horn, Obow and Clarinet feature some sustain sounds from Trumpet A. That does not sound right.

I would appreciate an answer/fixes.

Cheers, AKJ

Hi Kirk,

good to see you here. What about these bugs. Can you reproduce/verify? Will they get fixed?

[synchronizer]

See above.

There are only two explanations that I can come up with:

1- this is a bona fide velocity bug

2- when I compressed the files to NCW, something went wrong. There were a couple missing file dialogues that I ignored. Luckily I backed up the files before they were compressed. I can check that in a day or two if necessary. I'm pretty sure that the compression is fine though. The missing dialogues may have referred to some files that were simply renamed during the Ruby + Emerald = Diamond process.

Found the bug. I'll post the fixes tomorrow at http://www.kirkhunterstudios.com/upgrade.html

That's wonderful news! If you haven't checked other instruments, I'd do so though. I believe that the viola had the same sort of problem, among other instruments (I think the solo Romantic as well.) Maybe the bug, once fixed, repairs ALL affected instruments, in which case, there's no need to worry.

I'd like to add that I wasn't angry at you for the bug, but sincerely wanted to assist with the improvement of your products. I believe that I have done so. Once again, it's great that the problem has been found.

EDIT: Oh... and... I guess I won't be able to purchase Concert Strings II; ever...the funny thing is that I may have been able to afford CSII at $200 today.
Siigggghhhhhhhhhhhhhhhhhhhhh

[hyper.real]

Some benchmarks I have obtained with an SSD in two of my computers.

Background. Finding myself with CS2, CB2, Kontakt's libraries, and two licenses for Kontakt, I had the problem of running these on two computers. To cut a long story short, the cheapest and most versatile option I decided was to use an SSD to store the libraries. The cost of upgrading both machines to Windows 7 for its ability to access huge amounts of RAM, adding 16GB RAM, plus the time (and pain) of reinstalling existing software, was too much, especially as Kontakt implements DFD sample streaming from disk.

Tomshardware.com pointed me in the direction of Corsair's Force 3 Series, so a 120GB drive was procured. Corsair use the ATTO benchmark, so I used this also.

The two computers were (A) own build desktop with Asrock 4CoreVIIV mainboard ca. 2007, Core2Duo, SATA II supported, 2 x 200GB WD hard disks in RAID 1; (B) Thinkpad T400 ca 2008, Core2Duo, SATA II supported, 160GB hard disk.

(A) had free SATA ports to which the SSD could be directly attached internally. For (B) I obtained a drive caddy to go in the DVD/CD slot, to give me the corresponding direct connection to the SATA bus. For flexibility, and out of curiousity, I also obtained a couple of cheap 2.5" external drive enclosures, one of which claimed USB3.0 support, the other eSATA II (also USB2.0) support. To attach these to the T400, I obtained a couple of ExpressCard adaptors, respectively a USB3.0 card and an eSATA card. All these items were cheap offerings available from Amazon. All worked correctly, after obtaining the latest device drivers. In the process of testing I found the eSATA cable supplied with its enclosure was a limiting factor, so a better quality cable was used.

Because the SSD was specifically for sound libraries, I was interested in Read speeds; Write speeds I could care less. Also, I was less interested in absolute date transfer rates than in ensuring any particular mode of connection was operating to the max in these systems. Again, I am assuming Kontakt is sufficiently optimised in regard to DFD that reliance on DFD does not make operations CPU-bound. DFD is subject to disk access times, but in this respect SSDs are so far ahead of plattered drives that the case for using an SSD for storing sound libraries is one that makes itself.

Peak read transfer rates, then.

1. Installed system hard disks: (A) 40MB/s, (B) 70MB/s. Comment: I believe the slowness of (A) is due to an old-ish RAID controller and M/B support. Current hard drives can give >120MB/s.

(2) SSD with direct SATA attachment: (A) 285MB/s, (B) 285MB/s. Comment: this is about the max thruput to be expected from SATA II. The Corsair Force 3 drive itself can tranfer at approaching double this rate, but that requires (decent) SATA III support.

(3) SSD over eSATA: (A) 285MB/s, (B) 123MB/s. Comment: as the ASrock MB in (A) 'exports' a SATA connector as eSATA, the (A) result is expected. I am uncertain as to the (B) result; it could be the ExpressCard adaptor and/or slot interfacing, or it could be a driver issue. Either way, it appears the SATA connection has fallen back into SATA I mode.

(4) SSD over USB2.0: (A) 35MB/s, (B) 37MB/s.

(5) SSD over USB3.0: (A) not tested, (B) 74MB/s.

Conclusion: by far the best performance with an SSD will be obtained over a SATA or eSATA connection.

Further remarks: In general, what you will actually get from an SSD depends ultimately on the underlying bus - PCI, PCIe, PCIe 2.0. The SATA I, II, III standards correspond to these. The capabilities of USB3.0, often touted as offering bus rates comparable to SATA III are IMHO so far 'theoretical' only, and my results are commensurate with only a 50% saturated SATA I bus. USB2.0 connection does not merit the expense of an SSD.

As any benchmarking, this exercise can be criticised for the choice of its particular benchmark, and for the choice of any benchmark at all - which is to say, that benchmarking does not reflect 'real life'. I accept both criticisms. However, as alluded to, if you have to put up with DFD in Kontakt, then the case for a SSD rather than a mechanical drive is overwhelming. The exercise here is merely a demonstration of how this should be utilized. The hardware requirements for a Kontakt-based rendering of Mahler's 'Symphony of a Thousand' are not available AFAIK - if they were, we could then test better and worse ways of fulfilling them.

[KirkHunter]

Ok, go to http://www.kirkhunterstudios.com/update.html for the fixes.

[KirkHunter]

Hi,

so far, I found the following bugs in Diamond (yes, I have applied the update). Kirk, can you please test/verify and provide fixes:

1. VPRO Solo Brass: same bug as with Woodwinds (IntervaLive down does produce noise) where it was fixed. The VPRO instruments are important since some insturments are not present in the TVEC3 versions (i.e. trumpet a, trombone a and french horn a).
2. TVEC3 Woodwinds: English Horn, Obow and Clarinet feature some sustain sounds from Trumpet A. That does not sound right.

I would appreciate an answer/fixes.

Cheers, AKJ

Hi Kirk,

good to see you here. What about these bugs. Can you reproduce/verify? Will they get fixed?

Diamond solo brass a instruments are also fixed at http://www.kirkhunterstudios.com/update.html

[forbiddensilence]

Some benchmarks I have obtained with an SSD in two of my computers.

Background. Finding myself with CS2, CB2, Kontakt's libraries, and two licenses for Kontakt, I had the problem of running these on two computers. To cut a long story short, the cheapest and most versatile option I decided was to use an SSD to store the libraries. The cost of upgrading both machines to Windows 7 for its ability to access huge amounts of RAM, adding 16GB RAM, plus the time (and pain) of reinstalling existing software, was too much, especially as Kontakt implements DFD sample streaming from disk.

Tomshardware.com pointed me in the direction of Corsair's Force 3 Series, so a 120GB drive was procured. Corsair use the ATTO benchmark, so I used this also.

The two computers were (A) own build desktop with Asrock 4CoreVIIV mainboard ca. 2007, Core2Duo, SATA II supported, 2 x 200GB WD hard disks in RAID 1; (B) Thinkpad T400 ca 2008, Core2Duo, SATA II supported, 160GB hard disk.

(A) had free SATA ports to which the SSD could be directly attached internally. For (B) I obtained a drive caddy to go in the DVD/CD slot, to give me the corresponding direct connection to the SATA bus. For flexibility, and out of curiousity, I also obtained a couple of cheap 2.5" external drive enclosures, one of which claimed USB3.0 support, the other eSATA II (also USB2.0) support. To attach these to the T400, I obtained a couple of ExpressCard adaptors, respectively a USB3.0 card and an eSATA card. All these items were cheap offerings available from Amazon. All worked correctly, after obtaining the latest device drivers. In the process of testing I found the eSATA cable supplied with its enclosure was a limiting factor, so a better quality cable was used.

Because the SSD was specifically for sound libraries, I was interested in Read speeds; Write speeds I could care less. Also, I was less interested in absolute date transfer rates than in ensuring any particular mode of connection was operating to the max in these systems. Again, I am assuming Kontakt is sufficiently optimised in regard to DFD that reliance on DFD does not make operations CPU-bound. DFD is subject to disk access times, but in this respect SSDs are so far ahead of plattered drives that the case for using an SSD for storing sound libraries is one that makes itself.

Peak read transfer rates, then.

1. Installed system hard disks: (A) 40MB/s, (B) 70MB/s. Comment: I believe the slowness of (A) is due to an old-ish RAID controller and M/B support. Current hard drives can give >120MB/s.

(2) SSD with direct SATA attachment: (A) 285MB/s, (B) 285MB/s. Comment: this is about the max thruput to be expected from SATA II. The Corsair Force 3 drive itself can tranfer at approaching double this rate, but that requires (decent) SATA III support.

(3) SSD over eSATA: (A) 285MB/s, (B) 123MB/s. Comment: as the ASrock MB in (A) 'exports' a SATA connector as eSATA, the (A) result is expected. I am uncertain as to the (B) result; it could be the ExpressCard adaptor and/or slot interfacing, or it could be a driver issue. Either way, it appears the SATA connection has fallen back into SATA I mode.

(4) SSD over USB2.0: (A) 35MB/s, (B) 37MB/s.

(5) SSD over USB3.0: (A) not tested, (B) 74MB/s.

Conclusion: by far the best performance with an SSD will be obtained over a SATA or eSATA connection.

Further remarks: In general, what you will actually get from an SSD depends ultimately on the underlying bus - PCI, PCIe, PCIe 2.0. The SATA I, II, III standards correspond to these. The capabilities of USB3.0, often touted as offering bus rates comparable to SATA III are IMHO so far 'theoretical' only, and my results are commensurate with only a 50% saturated SATA I bus. USB2.0 connection does not merit the expense of an SSD.

As any benchmarking, this exercise can be criticised for the choice of its particular benchmark, and for the choice of any benchmark at all - which is to say, that benchmarking does not reflect 'real life'. I accept both criticisms. However, as alluded to, if you have to put up with DFD in Kontakt, then the case for a SSD rather than a mechanical drive is overwhelming. The exercise here is merely a demonstration of how this should be utilized. The hardware requirements for a Kontakt-based rendering of Mahler's 'Symphony of a Thousand' are not available AFAIK - if they were, we could then test better and worse ways of fulfilling them.

Thanks for the review of SSD performance...i've always pondered about getting one for my IBM T500...i assume you use the second bay slot for 2nd drive?
I got 8gb of ram..but i would i benefit from a ssd?
What was your feelings of the performance compared to a spindle hard drive,..is the gain only in how quick the instruments load up?
Thanks

[synchronizer]

words

Thank you very much! . I'm fairly certain that other instruments were affected as well, but I'd have to check.

EDIT: Al right, I ran some more tests, and as of right now, the solo cello is the only other instrument affected by the bug. Again, velocities 46-51 do not sound.

That's all of them as far as I know. The Viola and other strings seem fine.

[KirkHunter]

words

Thank you very much! . I'm fairly certain that other instruments were affected as well, but I'd have to check.

EDIT: Al right, I ran some more tests, and as of right now, the solo cello is the only other instrument affected by the bug. Again, velocities 46-51 do not sound.

That's all of them as far as I know. The Viola and other strings seem fine.

solo cellos fixed at http://www.kirkhunterstudios.com/update.html

[TheSteven]

Hi Kirk,

Do we need to download any of the December updates?

[TheSteven]

--- Correction & apology ---
Diamond TVEC updates are working fine - In my project/song the instance of Kontact still had the older version of nki file loaded.
Once I cleared it and reloaded it was fine.

...Steven

[DarkStar]

Hi Kirk,

Do we need to download any of the December updates?

I would, to be safe - check their Date Modifieds against what you currently have.


--------
Kirk,
Can I suggest that all of the Update download files are prefixed (or suffixed) with the product name, DIA_, CSII_, SS2_ etc

[Cruciform]

So is there a KH forum somewhere yet? Would like to particularly discuss how to get various sounds out of the samples.

[Forge Audio Designs]

So is there a KH forum somewhere yet? Would like to particularly discuss how to get various sounds out of the samples.

there's a few of us here!

https://www.facebook.com/groups/197101483701349/

[warrior545]

There are an online TVEC 3 User Guide here:
http://www.kirkhunterstudios.com/Concer ... rview.html
and
http://www.kirkhunterstudios.com/Concer ... rview.html


And "TVEC 3 At a Glance" guides here:
http://www.kirkhunterstudios.com/tvec3_strings.html
http://www.kirkhunterstudios.com/tvec3_brass.html

Are there any downloadable .html versions or .PDF versions?

[AKJ]

Hi,

so far, I found the following bugs in Diamond (yes, I have applied the update). Kirk, can you please test/verify and provide fixes:

1. VPRO Solo Brass: same bug as with Woodwinds (IntervaLive down does produce noise) where it was fixed. The VPRO instruments are important since some insturments are not present in the TVEC3 versions (i.e. trumpet a, trombone a and french horn a).
2. TVEC3 Woodwinds: English Horn, Obow and Clarinet feature some sustain sounds from Trumpet A. That does not sound right.

I would appreciate an answer/fixes.

Cheers, AKJ

Hi Kirk,

good to see you here. What about these bugs. Can you reproduce/verify? Will they get fixed?

Diamond solo brass a instruments are also fixed at http://www.kirkhunterstudios.com/update.html

thanks, that's great!!! But what about the solo brass b and the trumpet references in the woodwinds? is the latter intentional?

cheers, akj