MX-14.2 downloads

[Jerry3904]

From the Sourceforge server only, coming up on a month since release:

pae 7,136
non 2,722

TOT 9,858

The proportion of pae:non has risen from about 2:1 for MX-14 and 14.1 to now about 2.6:1

Also, the pae version is now being downloaded on Windows 50% of the time (pic)

[Jerry3904]

Now (data through 7/27)

pae 8,287
non 3,261

TOT 11,548

Very interesting is how flat the pae curve remains after the initial surge--good holding power that suggests strong multiplier effect (word of mouth, forums, blogs, etc.).

[joany]

Question: Leaving the amount of installed RAM aside, and assuming the CPU is pae-capable, are there any reasons not to install the pae version?

[chrispop99]

Question: Leaving the amount of installed RAM aside, and assuming the CPU is pae-capable, are there any reasons not to install the pae version?

There is a small performance hit when using PAE. How much is open to question, and on-line estimates tend to be in the range 1-10%.

I suppose the question really (within your parameters) is are there any reasons to use it.

Chris

[Jerry3904]

Coming up on 2 months since the point release (data through 8/17):

pae 11,062
non 4,116

Total 15,176

Curve for pae remains essentially flat at 1,000/week after the initial surge, 50% Windows machines.

[Jerry3904]

2 weeks later, still holding steady at over 1,200/week (SourceForge server only, data as of 8/31):

pae 12,729
non 4,806

Tot 17,535

pae: US 24%, Windows 47%
non: US 23%, Linux 47%

How can we interpret the huge diff in OS between pae and non? Conversion (from XP, for instance) versus preference (from another Xfce distro, for instance)?

[joany]

2 weeks later, still holding steady at over 1,200/week (SourceForge server only, data as of 8/31):

pae 12,729
non 4,806

Tot 17,535

pae: US 24%, Windows 47%
non: US 23%, Linux 47%

How can we interpret the huge diff in OS between pae and non? Conversion (from XP, for instance) versus preference (from another Xfce distro, for instance)?

If I interpret these results properly, 47% Windows pae downloads would imply 53% Linux pae downloads and 47% Linux non-pae downloads would imply 53% Windows non-pae downloads. I'd say a 47/53 vs a 53/47 split isn't very significant -- it's only a 6% bias. Or did I miss something in the math?

The fact that pae has over twice the number of downloads as the non-pae isn't surprising either, considering that a lot of computers in operation today have 4GB or more of RAM installed, which need pae to take full advantage of RAM. The slightly higher use of pae among Linux users could simply mean that Windows users don't know what pae is.

[kmathern]

...The slightly higher use of pae among Linux users could simply mean that Windows users don't know the difference.

That's kind of what I was thinking. Also when they see the word "non" they think they might be missing out on something if they use it.

[Jerry3904]

I suppose that's true...as long as Windows and Linux are the only two possibilities.

[Adrian]

I doubt it's a term confusion, we used plain language to make it clear what's recommended and why some people might need non-PAE:

Version: PAE (recommended)
Version: non-PAE (for older systems that don't support PAE)