As you noticed, some songs were been recorded a bit higher/lower than the standard tuning, either because the musicians chooses to calibrate their tuners differently or because of a record/playback speed variation. As a result, playing along those song gives a dissonant tone if you don't match the original recording.

 

I already modified some CDLC that didn't matched the original calibration, but I face an issue : with some songs, i'm stuck to 445 Hz even if I entered an higher value in RST (add arrangement/tuning pitch). I can't figure why, because I found an exception (Jamiroquai's Too Young To Die, set at 446 without hassle).

 

http://s21.postimg.org/qpn7nq9uv/Untitled_1.jpg

 

Any suggestion ?

If you're describing what I think you are, there was an update a while ago that fixed this. You may want to update to the latest version of the Toolkit.

If had to with the Rocksmith showing a slightly different tuning frequency that what was shown in the Toolkit.

If I've misinterpreted your problem, then please do clarify it for me :D

Well basically it's how RS interprets offsets that used for frequency, welp I'm sure most accurate way would be cent-based logic(not a frequency based) but as for now you could set a bit higher frequency to get rs to display pitch correctly.

It's not related to rstk 2.6.0.x bug, he's at one of beta versions, which is fine..

That's right Alex, the pic above is a screenshot from my computer. So i'm running RST 2.6.10.0-e2bfaacd beta ;)

 

I agree, cent-based logic would be easier to handle, but for now, we make do^^

 

So if i understand you well i'll try to put 447 or 448 Hz, to make the game show 446 ? Why not, i'll give it a try.

Already testing some changes tho http://i.imgur.com/O3wg4Fg.png

It looks like you're up to something ! ^^

 

Meanwhile i checked, and indeed it should be a discrepancy between RST and Rocksmith. Nevertheless, I assume the Rocksmith tuner is set to  exactly what asked on the*.pscarc. I experimented a bit, using decimals after the Hz value which change the cent accordingly, even if i cannot change it directly, it looks like works pretty much ok on the game, despite what it shows on the screen ;)

 

(I still can't figure why it shows - sometimes ! - 445 Hz when entered 446.00 Hz in RST, but... who cares, if it works in the end)

 

I still have to figure the exact tuning pitch of backing-tracks - i to that by ear for the time being, i guess there whould be a more accurate way to do that.

 

Thanks !

Jamiroquai - Too Young to Die is mine, you say A446?

 

I'll give that a try soon too, thanks :D

@@Jefbyos,@PC Plum,@@GetTheLedOut
You now can set exact cents value instead of playing around with frequency, it took long to implement\test for me, but it's there now :)

If you wonder why toolkit's fail to be even with Rocksmith, you can look at this data, I gathered all CentOffset values i could by that time i did it, and here small ref to what possible values can lead to what you'll see in Rocksmith:
https://docs.google.com/spreadsheets/d/1P8WyI0d1QqV0fc_G50zPvDw4d8yF5MU6knhQLGELJQ0/edit?usp=sharing

I picked your Latest Toolkit Build : Nice Work ! Thanks @@Alex360 ! Now i can directly enter the cent values i find using my tuner, to match the internal tuner of Rocksmith, and it seems to fit without hassle.

 

Apart that, i don't get your spreadsheet... It looks to be completely off, when entering 445 Hz you'll get 450.17 in Rocksmith, is that right ?

 

@PC Plum : I found +25 cent offset is working great - which makes 446.4 Hz on RST. But if you're not in a hurry, I'm also working to fix the Tempo map and when finished, I'll upload a V2 of Too Young To Die ;)

Since you mentioned Too Young I have been putting some thought into updating it myself. It was an early custom of mine and I am sure the tempo map could be imporved, but my intention would be to update a whole lot more than the tempo map.

 

Gimme a few days, I can make it beautiful :D

I almost finished the work on the tempo map... I propose that i polish my work tomorrow, an i send it to you so you can work on it. What do you think ? We'll both save some time ^^

TBH it would take me ages to get around to it, I got a little jelous and opened my big mouth :D

 

I'll struggle to find time to work on this, bash on man :D

I just noticed this post about a nifty little tool to detect the tuning pitch of songs.

 

Has anyone here used it extensively and can comment on its accuracy?

 

I tried it on my customs and it seems like it could be useful. For Vielleich Das Nächste Mal it gives 433.3 Hz which sounded too flat. Probably the slide in the track throws it off?

 

For Owaranai Uta it gives 438.6 Hz and for Money 441.7 Hz. Pretty close to 440 Hz, and nobody likes retuning... Is it even worth pitch-shifting them?

I tried pitch-shifting Owaranai Uta in Audacity to 440 Hz and the program gives 440.8 Hz for the resulting file.

Those kind of software will always be at least slightly false because of many component in sound making determining those things difficult with a fully mixed and mastered recording. And even the player might not have been perfectly tune in the first place (it's easy to be 1hz off even with a good tuner).

 

1 or 2 hz are such a slight difference that indicating the true tuning might not be worth it and won't be doable precisely on the guitar/bass anyway.

That's pretty much what I suspected.

 

The only A442 I have in my collection is Born to Run and that too is just the game showing A443 as A442.

I agree with Firekorn, guessing the pitch of a song can't be a perfect science, moreover involving acoustic instruments. Not only the tuning can be a bit off even if you take great care of it, but it does not stay as is during playing, strings can soften a bit, which can be easily becoming a nightmare if your guitar is equipped with a tremolo. But even if not, just the pressure of the fingers on the strings or the power of the attack with or without mediator, makes each note pretty much different to the others...

 

Add to that the fact that many instruments are involved (typically, one bass, one drum for which each element has it's own pitch), one or several guitars, one or several keyboards... Each one can have it's own inaccuracy, even synthesisers (at least, most vintage ones are known for that)... Interactions will lead each note to be slightly different to another.

 

(And we won't mention old recording techniques involving tapes, that induced Wow and Flutter measurement in each step from recording to mastering, and your playback as well on tape or vinyl).

 

Apart that, the Zplane tool you bring back to our attention should probably be helpful; i will definitely give it a try. So far for the few songs i wanted to correct, i used the pitch correction software in my DAW to isolate notes (tricking a bit with filters and gate to isolate the bass guitar more specifically), and then a tuner to have a measurement on most isolated ones i could find. Then, average the results, et voilà... Not the easiest way to get the job done, for sure.