Propellerhead Forum

Propellerhead Forum (https://www.propellerheads.se/forum/index.php)
-   Feature Suggestion Forum (https://www.propellerheads.se/forum/forumdisplay.php?f=7)
-   -   Performance Ratings for REs (https://www.propellerheads.se/forum/showthread.php?t=181182)

cpetersus 2013-04-23 13:51

Performance Ratings for REs
 
It would be nice to have a standard way that REs could be classified or rated based on their CPU usage. I'm not sure I'm phrasing this exactly how I want to but in general it would be nice to know how much of a relative impact an RE would make on a system. This could help folks make more practical decisions based on their system and reduce RE "remorse".

Additionally this may encourage developers to optimize their code if it will clearly be "rated". I understand that this is difficult as it really depends on features utilized in a specific preset. Perhaps there is a base performance rating and a maximum performance rating. Again, I'm not sure if I'm articulating my point very well but hopefully the intent is bring conveyed.

colcifer 2013-04-23 17:48

Many REs are cpu hogs because of the SDK. It might not be fair to the developers to judge them in this way. And who's to do the rating?

guitfnky 2013-04-23 19:08

if this could be done well, I think it's a brilliant idea. no one else does this sort of thing, and it would be very useful knowledge to have.

Philup 2013-04-23 20:01

As long as it was not left to the general public to do the rating. That being said, who will pay for the resources to do the testing and rating of each device?

cpetersus 2013-04-23 21:07

Its my understnding that Propellerhead does testing on ever RE prior to approval. If the testing is basic and standardized the rating would be relatively painless and unbiased. The efficiency (or lack of) the SDK is a matter of fact and somewhat superfluous. At the end of the day the RE rating would relate the RE to computer performance which is the goal.

Exowildebeest 2013-04-23 22:20

It's quite impossible to provide a consistent benchmark. I don't think you'll ever get more than a very rough distinction between "very cpu intensive - 4 cores recommended" and "light on CPU - you can use a lot of these simultaneously".

cpetersus 2013-04-23 23:43

I agree and really an indication Instead of a hard number would be more than acceptable for the purposes I'm referring to.

guitfnky 2013-04-26 21:08

Quote:

Originally Posted by Exowildebeest (Post 1294861)
It's quite impossible to provide a consistent benchmark. I don't think you'll ever get more than a very rough distinction between "very cpu intensive - 4 cores recommended" and "light on CPU - you can use a lot of these simultaneously".

definitely, but even that is better than nothing. :)

JensenTNI 2013-04-27 09:43

Its true that you could at least get a rough benchmark out as a guide. OK, you can of course test out all RE's but as with a review section, this could be still a useful and benefitial feature.

KarmaShaman 2013-04-27 10:16

As a new user of the more modern Reason, with plenty of new 'toys' to play with but a PC that suffers as a result, I think ratings would be an excellent idea. However, doesn't the 30 day evaluation period on RE's answer most people's questions ?

Although I suppose combining rack devices cant be done unless you trial all of them at once !

For what it's worth, overdriven guitar audio through the Echo, Polar, Trance Gate & pitch dropped bass track all at once is going to put a lot of computers to the test haha.


All times are GMT +2. The time now is 00:54.