Macbook 2.0 GHz or 2.16 GHz ? Ram ?
I am thinking about buying Macbook Pro. But I don’t know if CPU 2.0GHZ and 2.16 GHz really makes a lot of differences running Max/MSP. The 2.0GHz Macbook only provide 512 mb ram while the later one provides 1G ram. I was wondering if these two different ram sizes also make a lot of differences running Max/Msp for interactive pieces.
The model with 2.16 GHZ CPU and 1G ram is 400 dollars more expensive than the model with 2.0 GHz CPU and 512 Ram. But I only need the Macbook to run Max/MSP audio for me. Which model do you suggest ? Is Ram size that important for Max/MSP ?
Please give me some suggestions.
Thank you very much.
RAM size is important for any kind of computing. Totally worth it, IMHO.
(2 GHZ macbook pro, 2 gig ram)
Don’t buy the RAM at all, if it’s only MaxMsp you’re using. Even on my paltry Mac Mini 1.25Ghz, when I’ve got audio running at the limit and midi doing it’s best to keep up etc, I’m still only using a quarter of the 1GB that I put in, and that includes the system !
Processing speed yes – lots of it, as much as you can get (unless the extra 1.5 percent costs 500 quid), but MaxMap seems not to be interested in much RAM.
I dare say some will disagree.
It depends on what kind of stuff you’re doing in Max. If you use long
delay lines, loads of buffer~ you’ll need more RAM. On one project
I’m working on, the application uses more than 800 MB of Ram (only
because of multiple tapin~/tapout~)… if you uses Jitter it’s more
Anyway, on MacOS X the RAM is really important if you want to work
with decent conditions.
my 0.02 �
On 24 mai 06, at 11:42, Emmanuel Jourdan wrote:
> Anyway, on MacOS X the RAM is really important if you want to work
> with decent conditions.
And MacBook can use the RAM for its video card. And Rosetta needs RAM
Centre de Recherches et de Formation Musicales de Wallonie asbl
This is interesting…
"And MacBook can use the RAM for its video card. "
can you explain any more about this
Just read the small, light grey, characters on Apple’s website:
Graphics and Video Support:
Intel GMA 950 graphics processor with 64MB of DDR2 SDRAM shared with
main memory (1)
and the (1) footnote:
Memory available to Mac OS X may vary depending on graphics needs.
Minimum graphics memory usage is 80MB, resulting in 432MB of system
Centre de Recherches et de Formation Musicales de Wallonie asbl
I’m having a MBP 2.0 with 2 gigs of RAM and it is flying (with universal binary software on OS X). But as we know, Max MSP isn’t yet universal, so I tend to run it in the Windows environment via Boot Camp and it runs very well!
On 24 May 2006, at 13:54, Patrick Delges wrote:
> Intel GMA 950 graphics processor with 64MB of DDR2 SDRAM shared
> with main memory (1)
The MacBooks don’t have a separate graphics processor (unlike the
Pros) and has to use main RAM.
nick rothwell — composition, systems, performance — http://
I am wandering to purchase a Macbook either, do you mean a PC laptop is
enough for Max.MSP even for Jitter? thanks.
I mean that I use MacBook Pro with booting Windows to run MaxMSP at this moment since there is no Universal Binary version of MaxMSP. And it’s outperforming the most PC laptops. 256MB of OpenGL accelerated graphics is sure to satisfy everyone :)
It is interesting to read these entries. I live in Murcia, Spain. The last time I even SAW a Mac was during a visit to Hamburg. It is always humorous to me to come in contact with people and places where Macs are the standard. Because I give courses involving Max/MSP and PureData in a variety of settings, I’ve had the opportunity to observe performance on PC’s ranging from a 400 MgHz PII up to a dual-core Pentium D. I never bothered to set up a real test patch, but I can generally give some information about performance.
First: I have no doubt that you would generally be better off buying a Mac. They even look cooler. That having been said, however; here is the information about PC’s:
I use two PC notebooks daily, a P4 with 3 GHz and a Centrino with 1.70GHz. For both of them, I use cheap, external USB (1.0) audio devices, although for more detailed work I use a firewire device. (The USB devices are small and hand, the firewire large and heavy. This is due to the inherently low quality of onboard audio-chips; they have high latencys. How is it in the Mac community?) The centrino is, of course, more limited than the P4, but while processing audio, I still haven’t managed to make it peak. The limits are much more apparent with video, which demands better graphic-cards than notebooks will give you, anyway.
RAM is important, but only if you are doing things which force you to go beyond the RAM you have. Buffers (alone, in stutter, tapin, delay or any other object) use up exactly the amount of RAM you assign to them, so you can figure out how much you need. For all of the work I have done, 500 MB will suffice (and do so nicely), but 1 GB allows more headroom. Figure it out: a CD has 74 minutes of stereo information in 650 MB of space. This means 4.39 MB per minute of mono in 44100, 16-bit resolution. My computer needs about 150 MB of memory before it is ready to do anything, meaning that I can (with 500), use about 79 minutes worth of buffer space before my computer starts to have to write things to disc. Do you need more than that? In my larger projects, I load sounds as the performance goes along in order to allow more than my RAM will take.
CPU power is nice, but doesn’t make itself as noticeable as one might think. Only with multiple FFT-objects does it begin to play a role. On an old 700 MHz machine at our theater, I can happily use 3 gizmos at once, transposing without glitches, as well as process the sound in other ways (delays, loops and the like), but that about tops it out. Don’t get too hung up about speed; we’ve been doing fine in the audio world since they started selling 300 MHz machines. Believe me; my first PC was a 70 MHz 486, and a friend of mine is STILL making CD’s with it, using a digidesign session8.
The bottom line is this: any notebook, PC or Mac, will give you good audio performance if you configure it correctly. The performance differences will come into play when you need to duplicate what you are doing, such as transposing input 10 times instead of 3, or having 100 1-minute buffers instead of a few 20-second buffers or a collection of 2-second buffers. For me, the limit is reached when I want to use a single machine to do all the things I am doing, instead of chaining various machines together. Since I do alot with video-tracking, I need one computer for the tracking, one for the music and (not always) one for live-video. I need to experiment more with dual-core machines, but the present, multiple-computer approach works well.
Hope this helps.