Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Sun Baked said:
That's your choice. :(

Especially when there is a lot of evidence to support the fact the bridge controller was a PCI/PCI-X HT Tunnel.

The low end original G5s did not have the PCI-X bridge chip, it only had the 64b/33MHz slots and those hung off of the K2 like you see in the current PowerMac Dual G5 1.8GHz. I believe the current Dual 1.8GHz is the same chip set as seen in the prior low end system, only the CPU is different.
 
Hey that's a pretty cool way that Apple supports standard PCI on the low-end mobo. The things HT makes easy. :)
 
shawnce said:
The low end original G5s did not have the PCI-X bridge chip, it only had the 64b/33MHz slots and those hung off of the K2 like you see in the current PowerMac Dual G5 1.8GHz. I believe the current Dual 1.8GHz is the same chip set as seen in the prior low end system, only the CPU is different.
Methinks all you guys are wrong, and so does Apple. ;)

But I'll wait for you to prove both me and Apple wrong.
Block Diagram on Left
Apple dev note: Power Mac G5 (Jun 04) said:
PCI Only

Some configurations of the Power Mac G5 support three 64-bit 33 MHz PCI slots that interface to the K2 I/O. The PCI configuration conforms to the PCI Specification 2.3.

PCI-X Only

Some configurations of the Power Mac G5 support three PCI-X slots that interface to the HyperTransport bus via the PCI-X bridge. One slot runs at a maximum of 133 MHz and two slots run at a maximum of 100 MHz. The 133 MHz slot can support a maximum burst bandwidth of 1064 MBps, based on 64 bits times 133 MHz. The two 100 MHz slots can support a combined bandwidth of 800 MBps. It is recommended that the highest bandwidth card be inserted in slot 4.
Block Diagram on Right
Apple dev note: Power Mac G5 (Aug 03) said:
PCI or PCI-X Expansion Slots

The Power Mac G5 provides three PCI or PCI-X slots via the HyperTransport bus. In some configurations of the Power Mac G5, one slot runs at a maximum of 133 MHz and two PCI-X slots run at a maximum of 100 MHz. Other configurations of the Power Mac G5 computer access all three PCI slots at 33 MHz. Each slot has room for a full size 12.335-inch or short 6.926-inch card.
The Picture with the Block Diagrams
 
I don't doubt you...

I donot doubt that this is true - but my order date is still marked on or before 7/30 - although I was probably order number 3 or 4... maybe I get one of the first cards?

NEW orders do switch to Augues ship date and the card alone states late Auguest...

So I won't be dissapointed... yet I'll remain optimistic.... "On or before Aug...uh I mean JULY 30th"

IJ

Freg3000 said:
When you add the 6800 to a Power Mac for any model the ship date changes to August.
 
gekko513 said:
Howcome you get to know what Apple thinks and I don't :p
Well you guys were "reading" the block diagrams instead of looking at the machines architecture. ;)

Apple tends to just show the block diagrams will all the goodies, when the low end of the line may not end up without everything in the picture -- or use PCI instead of PCI-X.

But you have to actually look at the block diagrams to the machines sometimes, and disregard the consumer specs, to see what games Apple is playing.

They may look the same on the outside to the consumer (feature wise), but be vastly different on the inside. So while SDR eMac looked like a CRT version of the LCD iMac to the consumer, it actually looked a lot more like a PowerMac G4 when you looked at the block diagram.
 
has someone collected these block diagrams into one centralized place where it would be easy to compare, or do they have to be searched one by one from apple site?
 
Sun Baked said:
Methinks all you guys are wrong, and so does Apple. ;)

But I'll wait for you to prove both me and Apple wrong.The Picture with the Block Diagrams

You mean the single processor 1.6GHz machines really had 2
CPU's and 8 Dimm slots - where did they hide them? On the back
of the board? I mean they must be there - the block diagram shows
them clear as day, it must be true. ;)

Clearly the 'old' block diagram has a few "extras" compaired to the
real single 1.6GHz G5. If they include 'little' extra's like a CPU and
memory it wouldn't surprise me if the PCI-X bridge was an extra for
the 1.6.

Given they are all talking HT it makes perfect sense to me to skip the
PCI-X bridge when the machine doesn't have PCI-X. This makes as much
sense as skipping the extra processor when the machine doesn't have
two CPU's.
 
Dual 2.5s w/6800 in July??!?!?!

For those of you keeping score at home, here's my order status:

06/28/2004
11:52 AM PST
APPLE CIN HD DISPLAY 23" FLAT PANEL-USA
Open Est. Ship 07/26/2004
2.5DPG5 CTO
Open Est. Ship 08/31/2004

So you can see I ordered the day of the announcement, and I'm listed as the end of August. I guess my trusty B&W G3 will get to taste the silvery goodness of my 23" Cinema for a month before big mac daddy arrives. As with the others, I'm hoping he arrives before the end of August!


iriejedi said:
I donot doubt that this is true - but my order date is still marked on or before 7/30 - although I was probably order number 3 or 4... maybe I get one of the first cards?

NEW orders do switch to Augues ship date and the card alone states late Auguest...

So I won't be dissapointed... yet I'll remain optimistic.... "On or before Aug 30th"

IJ
 
2.5Ghz XServe

joeboy_45101 said:
I wonder when we'll see a 2.5Ghz Xserve??? That would be a very kickass machine for cluster computing.

Your joking right!

I've had 5-Dual Processor 2.0Ghz Xservers on order for 3 months and I have not received them yet. The dealer says they have finally shipped and I should receive them later this week or next week.

Don't expect them anytime soon, I wouldn’t expect to see them until later this year or early next year.

Although it is a nice thought to get them earlier, with my luck Apple will announce them 2 to 4 weeks after I get my Dual 2Ghz XServers and with a price reduction as well. LOL :)
 
EarlMacT said:
Your joking right!

I've had 5-Dual Processor 2.0Ghz Xservers on order for 3 months and I have not received them yet. The dealer says they have finally shipped and I should receive them later this week or next week.

Don't expect them anytime soon, I wouldn’t expect to see them until later this year or early next year.

Although it is a nice thought to get them earlier, with my luck Apple will announce them 2 to 4 weeks after I get my Dual 2Ghz XServers and with a price reduction as well. LOL :)

Just don't get me started again ;)

You're a lucky one! I ordered mine 7 MONTHS ago, and am still waiting :(

Current EXPECTED ship date is August 3rd and has been moved back each time it hit it's expected ship date since March.

So take ALL Apple EXPECTED ship dates with a very LARGE grain of salt!

BTW, I have a HW RAID card in mine. Without it, I just don't need the server.

MM
 
My 23 inch display

Apple just ran the charge for my new alumn 23" display.

See, how it works is, Apple usually charges you for the product just before they ship.. So, maybe, and I do mean maybe, this implies that my display will ship within the next couple of days?!
 
thedogcow said:
Apple just ran the charge for my new alumn 23" display.

See, how it works is, Apple usually charges you for the product just before they ship.. So, maybe, and I do mean maybe, this implies that my display will ship within the next couple of days?!


You're absolutely right. Actually they don't charge until the day of shipping in my experience. Looks like you've got a 23" on its way! :cool:

I ordered my 20" the day of announcement. When did you order yours?
 
CPU

To get back on topic, my 2.5GHz cpu ship date says on or before 7/30.

However, I ordered 3 things ( 1) adc-vga adapter 2) 23" display 3) the CPU)... 1 of that has already shipped and 1 that is probably going to ship very soon as to elluded above. Hopefully, the cpu will arrive before the 30th.
 
Sun Baked said:
Well you guys were "reading" the block diagrams instead of looking at the machines architecture. ;)

Actually I am looking at the system/architecture (not just the block diagrams) but don't let that get in your way. As I said a PCI-X bridge is not used in the original low end PM G5.
 
thedogcow said:
Apple just ran the charge for my new alumn 23" display.

See, how it works is, Apple usually charges you for the product just before they ship.. So, maybe, and I do mean maybe, this implies that my display will ship within the next couple of days?!

It is normal business practice to charge your credit card the day of shipping. Have you checked the Apple site for your tracking number? You should be able to check on your delivery date very soon.
 
shawnce said:
Actually I am looking at the system/architecture (not just the block diagrams) but don't let that get in your way. As I said a PCI-X bridge is not used in the original low end PM G5.
Which is why the actual Apple Docs say "PCI or PCI-X slots via the HyperTransport bus," about the the original low end PM G5.

But if you guys wish to think otherwise, I'll let you from now on, even though it's not true.

Much easier to cripple the 100/133 MHz operation with a few resistors and/or a couple lines of code and use the same circuit board, like they did.

The PCI-X bridge chip is capable of it, and Apple does like to do this stuff.
sample bridge chip docs said:
If the systemboard supports PCI-X mode operation for a bridge, then a pullup resistor to VDD33 must be placed on the bridge’s PCIXCAP pin. To limit the frequency of a PCI-X-capable bridge to 66 MHz on a systemboard, the systemboard must also include a pulldown resistor from the bridge’s PCIXCAP pin to ground. The strapping options on GNT[4:3]# are used to distinguish between systems that support 100MHz and 133 MHz; in either of these two cases, the system board should include no pulldown resistors on PCIXCAP.
But that's not possible in your fantasy is it?
 
dontmatter said:
weird... not that I know enough to interpret any of this, but it SEEMS like the old version... the PCI stuff, as you said, and also half the space for memory....

???

wouldn't the old one be better, by these pictures?
Basically the old DP 1.8 G5 PM takes advantage of the PCI-X speeds in you ever need to add in an upgrade card and it supports it.

The old low-end SP 1.6 machine had the capability to overcome it if the overclockers/hackers worked on it (like they did for monitor spanning on the iBooks), but these new low-end machines most likely not be able to based on this change.

All depends on what the capabilities of this new updated KeyLargo2 are, and it's much harder to figure out since there are no docs/technical references in the public domain like there are for the many HT PCI-X Tunnel chips.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.