Traktor 2.6 Hogging CPU!

[No Left Turn]

@GeekGod... interesting you say that... because I've worked with developers before and they always ask for screenshots or screencasts of certain reported issues. In some cases it makes sense, because if you're making something happen it'll show in whatever workflow you're doing, but it's stupid/rude/disrespectful to ask for it as "proof" of it actually happening.

[Scaper7]

What's really funny is NI wants me to prove this with a video. They do not believe their own customers. Apparently the developers in 'berlin' are really hard to get along with from the bits and pieces that I can gather from USA support and require 'proof' of issues. Silly.
Sucks we are caught in the middle but I cannot be bothered to 'prove' there is an issue with their software.

Doesn't surprise me one little bit. After two years as a NI customer I've learned not to waste my time with NI support. Acknowledging something might be wrong with a product is just not part of their brief. They have sent me on ridiculous wild goose chases, wasted my time and money to buy items they claimed would fix my problem but did nothing.

With regards CPU hogging, I've just bitten the bullet and bought a new quadcore i7 laptop (Win7) so I can just get on with using Traktor, despite CPU load issues. My Core2Duo was glitching occasionally with 2.5/2.6 but not on the new i7 system. 0-1% CPU load with Traktor sitting idle (FX on, Keylocks on) ... even with 2 tracks running, load is negligible (still only around 1% registering in task manager/performance tab, occasionally climbing to around 10% briefly, then back 0-1%).

[Rudi-J]

... but not on the new i7 system. 0-1% CPU load with Traktor sitting idle (FX on, Keylocks on) ... even with 2 tracks running, load is negligible (still only around 1% registering in task manager/performance tab, occasionally climbing to around 10% briefly, then back 0-1%).

Good for you it improved with the new hardware....
Please do me a favour and check what happens to your TP2.6 CPU load control display when you raplid twist the knob/encoder adjusting loop length or if you rapidly press hot-cue buttons?

[Nicky H]

Good for you it improved with the new hardware....
Please do me a favour and check what happens to your TP2.6 CPU load control display when you raplid twist the knob/encoder adjusting loop length or if you rapidly press hot-cue buttons?

I've tried this and nothing...
My pc is a few years old now;
intel i5 2.8
8gig ram

4 decks with macro fx on is about 9% in task manager

macbook still on 2.5, haven't updated because of all the problems listed in here but pc seems ok

[Rudi-J]

4 decks with macro fx on is about 9% in task manager

So the PC is on 2.6 and the CPU load display of traktor remains in green (not talking about the task manager)

[Nicky H]

Yes mate, no problems at all...
Sorry for bad screenshot:
ss.jpg

p.s yes that is 2 Unlimited in deck a

[Rudi-J]

Thanks for double checking Nicky,
I am now in contact with NI service - hope they have an idea (at first they wanted me to unload Macro FX and release the FX link to the decks, but this seems to have nothing to do with it)

[Rudi-J]

Just made another interesting observation:
Twisting loop length encoder will push up CPU load no matter what decks, but if deck c for instance is set to remix deck (and of course also when it is set to "live in") the CPU load remains stable. If the remix deck is thus loaded with 4 samples then the loop on deck c will play fine, can also be rapidly adjusted in length and position by encoders without clicks and pops and all this with CPU load in traktor display not exceeding 50%.

So the bug eats up CPU time only for track decks....

[Nicky H]

Weird...
Even weirder that it affects some machines and not others..
Some mac, and some pc..

Has to be a Traktor bug surely?

[GeekGod]

That would be my guess. I sent over screen shot "proof" that the problem exists and NI is now trying to get the programmers to understand there is an issue. The way I see it is a tug of war between support and the developer in berlin. For some strange reason I get the feeling the programmers have their head up their anus and don't want to believe anything support says.