| How Many Windows Can You Open On the Natami? | |
|---|
|
|---|
Chuck T
USA
| |
Posts 679
17 Jan 2012 15:07
| How many windows can you open up on the Natami? I propose you write a program to open up a window, run an adding program to count from 1 to infinity, open another window to do the same thing and count the number of windows of programs running. Run it until it crashes and then report with the final number.
| |
Thomas Richter
Germany
| | (MX-Board Owner)
Posts 1425
17 Jan 2012 18:59
| Chuck T wrote:
|
How many windows can you open up on the Natami? I propose you write a program to open up a window, run an adding program to count from 1 to infinity, open another window to do the same thing and count the number of windows of programs running. Run it until it crashes and then report with the final number.
|
There is no definite answer to this question since it depends on many other factors as well. Which type of window, which type of refresh handling? If you have a smartrefresh window, chipmem for the layers buffer is the limiting factor. If a single program opens the window, and does not share ports, the number is limited by the number of signals available for the message ports of the window (e.g. at most 16 windows). The window itself will not require much memory, but many other resources - I assume that layers and as such intuition will slow down to a crawl and become unusable before the system runs low on memory.
| |
Chuck T
USA
| |
Posts 679
17 Jan 2012 22:54
| I think it would be a good benchmark test to test against a stock Amiga.
| |
Phil "meynaf" G.
France
| | (Natami Team)
Posts 393
19 Jan 2012 08:14
| I have opened over 200 cli windows simultaneously on my A1200 without experiencing any problem. I think it's more a matter of available memory than anything else ; not a very good benchmark in my taste.
| |
Chuck T
USA
| |
Posts 679
19 Jan 2012 20:41
| How do you introduce new people to the Natami if they don't know what it can do? This is how they sold the old machine so it might not be a very good benchmark but an important demonstration none the less.
| |
Lorenzo Vivarelli
Italy
| |
Posts 9
22 Jan 2012 19:40
| In my opinion, "an important demonstration?"
It is the opposite. I beg your pardon.
| |
Fabian Nunez
USA
| |
Posts 312
23 Jan 2012 03:17
| This tests the OS and not the hardware. I'd expect a Classic with 512MB of RAM to perform pretty much exactly the same as a Natami on this particular test.
| |
Jakob Eriksson
Sweden
| | (Moderator)
Posts 1097
23 Jan 2012 07:38
| Yes, me too.
| |
Ceti 331
United Kingdom
| |
Posts 282
23 Jan 2012 16:43
| Agree this sounds like a very convoluted test which is more an OS/software design issue than a hardware issue, no parameters defined like size of windows etc. but perhaps you are wanting to benchmark the efficiency of multitasking ? .... there's useful statistics there, e.g. at what granularity threading/multitasking is useful.
This might differ a lot between different OS/hardware approaches, especially the MMU vs non MMU.
experience with game engines is minimal use of threading, just one per core, and everything managed through job queues etc - to eliminate OS thread swap altogether.
apple GCD is a bit like that i think.
| |
Casey R Williams
USA
| |
Posts 149
30 Jan 2012 03:46
| I am not so sure I see the point here either. What is the maximum number of windows you would ever use at one time? What is it a benchmark of exactly? It almost seems like a low limit (maybe 10) would have been part of the OS design. I am almost surprised it is not.
| |
|
The team will post updates and news here