A comparision with last month's NPD (adjusted for the difference in weeks) gives this:

Wii ...... +33%
DS ......... -7%
PS2 ...... +4%
PSP .... +11%
X360 .... +8%
PS3 .. +108%
GBA ....... -1%

@turin: they do; it won't happen immediately, and it won't match NPD exactly, but corrections will come.

The message was more directed toward Luzillo, Diomedes1976 ;)  I asked because not a lot of people understands spanish, and I include myself in it, I took one course 6 years ago and I can understand basic spanish, but everything was not clear in that ;)  Thanks for the translation, and I'm happy to see that you understood my last sentence, it's nice to see some people who can speak French ^^

 And about the numbers................................Could not be happier :P  I can't even start to imagine what it will look like with MP3 in America, and SMG and SSBB in 3 months!

Absolutely, turin. If it becomes clear that figures are off, they are adjusted. This means that monthly totals may vary by 5-15 percent, but the lifetime totals are quite accurate. In fact, look here:

http://vgchartz.com/forum/thread.php?id=5078

As of the end of June 2007, Sony's official tracking of PS3 sales was less than 1 percent different than VGChartz' for it's lifetime. 

In other words, the monthly sales are pretty accurate. 5-15 percent off is the norm now. The lifetime sales are very accurate. 

NPD is a reliable source.

But look at the worldwide breakdowns:

Wii gets ~35-40% of sales from the Americas.

360 gets ~60-65% of sales from the Americas.

PS3 gets ~40-45% of sales from the Americas.

That means for July the 'discrpency between professional tracking agencies' breaks down like this:

PS3 was 3% different from NPD.  .40 x .03 = 1.2% difference - not worth changing given NPD's sampling has roughly 3-5% margin of error by my estimates.  Overall - this is a 7k/4200k difference!  Not a big deal (1/10 of 1% adjustment if it was adjusted for overall totals).

For simplicity, say Wii was 21% different from NPD in July.  .21 x 3/8 = 7.875% difference.  Might be worth bumping up a bit, but the figures are still very close worldwide 85k/10570k (8/10 of 1% adjustment if it was adjusted for overall totals) 

360 -  .26 x .625 = 16.25%.   Again, the totals are close worldwide 47k/10520k (4/10 of 1% adjustment if it was adjusted for overall totals)

In other words, there may not be a need for a bump if ioi sees it as I do. 

Final gif for the night ...

Ok Thanks to KruzeS and Bodhesatva, if a console's lifetime sales are accurate, then that is what matters to me, now i can trust in Vgchartz.

Realistically speaking, the accuracy of NPD is probably in the range of (+/-)2.5% 19 times out of 20 whereas vgcharts is probably in the range of(+/-) 10% 9 times out of 10 ... What this means is that vgcharts will always end up with a few consoles which are quite a bit different than NPDs data but (on the whole) should get somewhat close and be very usefull in determining the approximate ratio of how two consoles are selling or how games are selling.

Honestly ioi doesn't have the resources to track data like NPD does so you're always going to have to accept a certain ammount of error ...

The fact is that this really shouldn't matter all that much, most people are hear to discuss more about how consoles are selling in comparison to each other (and historical sales) and the charts ioi gives us are plenty good enough.

I'm a little sick of VGC apologists. You can torture the numbers however you want, but the fact is, there were a lot of horrible estimates 6 months ago, and a lot of horrible estimates today.

For a while now, I've been mostly ignoring VGC's weekly NA figures. Almost every feature on this site is awesome, but the weekly numbers remain worthless for all but the most casual observation.

Usually I enjoy your posts, but this statement is kind of skewed.  You are comparing Handhelds plus consoles to consoles only.  Nintendo still owns supreme, though.