• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Spaces on PCBs? Why?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

>HyperlogiK<

Member
Joined
Nov 10, 2004
Location
Sword Base
Why are there often so many empty spaces on the PCBs of mobos, gfx cards, sound cards etc? I realise that, for example in comparing an NF7 and an NF7-S the NF7 has a missing space where the SATA controller is on the NF7-S (and similar examples exist of cards with less RAM, no TV-out, no Tuner, no optical-in etc than the top model, but sharing a common PCB), but often even the top board or part in a series has many missing spaces with contacts (obviously originally intended to solder components to). Surely this must be un-economic? why does my DFI board (which as far as I know was the top Socket A DFI board) have so many spaces? If it is the top board then this isn't to save money by making a shared PCB for several designs. Surely considering the vast numbers in which many boards and other components are manufactured, to make a larger PCB with extra tracks and contacts that are never used will be a big waste of money?!
 
with video cards, they use the same PCB for the 128mb and 256mb versions, so they just don't put in the extra memory on the lower version. Same with Mobos, they use the same PCB from version to version say the Socket A Lanparty Ultra As and Bs, they have different audio and internet add ones that may need different componets like caps and resistors, so they leave those spots empty depending on the need
 
even the top featured cards with the most mem avaliable often have gaps though. Are you saying that this is perhaps anticipating a revision with even more mem, and so maybe more caps, or like the Voodoo 3 where it was originally intended to have more memory than it ended up with?
 
It's like computer programming, you first come up with a working solution. But then after looking at it you find ways to make smaller, tighter, more efficient code.
Electrical Enginering is the same way. They find ways to atcheive the same circuit using fewer parts. But they don't want to throw away the thousands of circuit boards they already have. So you have fewer parts on the board and Rev1.0 becomes Rev1.1
 
So not only empty space are left for high end hardware and future revisions but also appears on later revision as new technology comes along to allow tighter integrated design.
 
the boards are multi-layered. are you sure theres nothing in that gap? take a drill and find out ;)

bxp58220.jpg

from http://www.fotosearch.com/BDX320/bxp58220/
 
Last edited:
Quick, who has tar and feather??? Someone posted a picture of an ancient AT board with ISA slots! Such posts belongs to OG forum! (Old Geezers)

:p j/k
 
If it weren't for the OGs that frequent the OG forums, you young Geeks wouldn't even exist.....LOL
 
i'm guessing but i would assume that they would want to be able to accomodate as many revisions of a product as possible into a single manufacturing line as this would have large cost benefits by not having to change the design
 
Another possibility for unused solder points is their use for debugging purposes. When one designs a PCB, I'd immagine that they'd design in various test points which can be used in dubugging a malfunctioning board.

JigPu
 
Its easier to make ONE board for a whole line, and then just add different components to them.
 
eab said:
I got a quick qustion, I grew up playing on motherboards with ISA slots and the rows on rows of memory slots... does that make me an "Old Gesier"? I am only 19 flipping years old!

19 goin' on 90! :D
 
Maybe the cost of engineering separate PCBs for each exact model outweighs the cost of wasting some PCB space by using just a single design?
 
Back