Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Success: 2006 Mac Pro upgraded - 2 x Xeon 5355 Quad-cores!

Neodym said:
That's not quite right :)

----------


Iirc that is a mod (BSEL) to run cheaper 5345's with 2.33GHz instead of the regular 1.83GHz those chips are being advertised for. Nowadays the price of the 5355's is not that far off anymore, so you don't need to go 5345 and perform that mod (unless you want to have the cheapest upgrade possible or want to try to overclock the 5355 ;)
Click to expand...

thx, 5355's are in the mail. Comp is pretty snappy with the ssd - guess the ram difference is hard to notice (14 gigs now). The 5355's were $58 each - they cant be overclocked. Would have to go with the 5365's which are double in price, so ill just be happy with 4 mores cores. 4 MORE CORES!
 
Out of curiosity...

Which ones are the 2.4?

I see Geekbenches of MP1,1s with 8 Cores at 2400 that are higher than some of the 2.66 scores

??????????

Also.... what else can be done to bump the GB Scores?
What all effects the score?

I see scores up and down on these things, and see some 8 Core 2.66 scores higher than mine.

No fair! :(:eek::p
 
TableSyrup said:
Out of curiosity...

Which ones are the 2.4?

I see Geekbenches of MP1,1s with 8 Cores at 2400 that are higher than some of the 2.66 scores

??????????

Also.... what else can be done to bump the GB Scores?
What all effects the score?

I see scores up and down on these things, and see some 8 Core 2.66 scores higher than mine.

No fair! :(:eek::p
Click to expand...

think that would be the 5340's maybe? the 5355 is 2.66 and the 5365 is 3.0
 
5365's are 3.0's listed as 3000
5355's are 2.66 listed as 2660

I assume the 2400 it lists is a 2.4 GHz????
I'm seeing 2400 listed with higher results than some 2660

Anyway, I guess my more important question is what else impacts the GB scores

Cause I see a lot of variation even between the same processors (Respective to 32/64)

Even some 2.4(?) scores even rank higher than some 2.66
 
TableSyrup said:
5365's are 3.0's listed as 3000
5355's are 2.66 listed as 2660

I assume the 2400 it lists is a 2.4 GHz????
I'm seeing 2400 listed with higher results than some 2660

Anyway, I guess my more important question is what else impacts the GB scores

Cause I see a lot of variation even between the same processors (Respective to 32/64)

Even some 2.4(?) scores even rank higher than some 2.66
Click to expand...

ram and video card somewhat, mostly cpu. I dont know your history - is it possible that one of the cpu's is not installed correctly?
 
I've built around 50-60 computers.... I'd be surprised if I botched that up... would be a first... first for everything, but I doubt it. Pretty fool proof ;)

Everything got a nice thorough cleaning, no dust or anything in sockets or on CPUs before going in. Same technique for compound application I've always used without issue.

My CPU's were purchased locally - NOS from a reputable server company (Which is why I paid a little more... I think $150 ???)

I'm not out of the norm in my GB Scores... just wondering what else effected.
I've seen people writing that the CPU is the only factor - but I found that highly suspicious based on the tasks the GB testing performs.

I still have a 7300GT in there... and I suspect that perhaps an upgrade there will help???

RAM slots and chips, as well as riser cards and slots, were cleaned prior to installation as well...
16 GB ... 8x2GB (New as well)

Everything is running just fine, and I am happy with the machine,
it's just like I said, I saw higher scores and was curious why ;) That's all

I suppose I'll find out if the GPU helps soon, as that is my next upgrade ;)

Lowest 32 Bit GB I had so far was a 9300 or 9400 something. Not out of the norm for this config.

Thanx ;)
 
TableSyrup said:
I guess my more important question is what else impacts the GB scores

Cause I see a lot of variation even between the same processors (Respective to 32/64)

Even some 2.4(?) scores even rank higher than some 2.66
Click to expand...
Only CPU and RAM are tested.

Possible impacts on CPU part: Running background tasks, used OS version.

Possible impacts on Ram part: Running background tasks (if Ram is full and swapping starts), used OS version and used memory sticks on some machines. On the 1,1 for example it is best to have 4 Ram modules of same size. That way you have quad channel access enabled, which offsets the higher latency from 4 modules instead of only 2.

If you use more modules, latency increases (inherent to FB-DIMM). If you have different sizes, quad channel access is disabled.

In every day usage those differences are usually small enough to ignore and go for more Ram or different sizes for sheer cost reasons. In a benchmark however you'll probably see a difference.

Of course other factors may in theory affect results, like e.g. flaky hardware or overheating conditions, but I assumed a fault-free system here.
 
JesterJJZ said:
This is what I want right?

http://www.ebay.com/itm/SLAEG-Intel...Processors&hash=item1e6f86dbc7#ht_2284wt_1154

Seems like a good deal no?

Going from Original Dual Core 2.66, worth it?
Click to expand...
Wow - those have come down in price a long way! :eek:

If its worth it depends on the software you are using and whether it scales with more CPU cores. Handbrake for example will definitely show a vast improvement, more recent Adobe software as well. But others may still not be using more than 1-2 cores and then you will only see improvements under heavy multitasking conditions...
 
Neodym said:
Wow - those have come down in price a long way! :eek:

If its worth it depends on the software you are using and whether it scales with more CPU cores. Handbrake for example will definitely show a vast improvement, more recent Adobe software as well. But others may still not be using more than 1-2 cores and then you will only see improvements under heavy multitasking conditions...
Click to expand...

AE, FCP, Photoshop mostly.
 
Mp 2,2

Can I upgrade to an intel cpu?


Model Name: Power Mac G5
Model Identifier: PowerMac7,3
Processor Name: PowerPC G5 (2.2)
Processor Speed: 2 GHz
Number Of CPUs: 2
L2 Cache (per CPU): 512 KB
Memory: 2.5 GB
Bus Speed: 1 GHz
Boot ROM Version: 5.1.8f7
Serial Number (system): RM428ZHSQPM
Hardware UUID: 00000000-0000-1000-8000-000A95D4C0D2
 
Didn't test my system for geek bench score prior to the upgrade but afterwards I scored a 93xx. Upgraded to 2 - x5355 from eBay. Seller was selling a lot of them for 58 or best offer so I offered 100 total and s/he took it. Amazing that I have a desktop with a higher geek bench score than some 2009 Mac pros and faster than any Mac Mini, but cost me less than the base Mac Mini. I needed a machine that could handle a lot of VMs at once and now I have it. When apple upgrades the Mac Pro next year I'll go new then but for now this will tide me over!
 
Just ordered! Got 2x 5355s for $100!

I'm tempted to get more RAM. I have 16GB with 2GB so I'd need to replace all of it if I wanted 32GB. I see deals for about $600-$700 to get it, plus I can probably sell my 16GB for $150 or so. Trying to figure out if I should or not.
 
Just finished upgrading, from quad 2.66ghz to octo 2.66 ghz. Had some trouble with the fan assembly, specifically the second time around when the fan didn't power on (wire moved slightly). Upgraded to the 2,1 firmware, and all seems to be fine and dandy.

For the 2x x5355s, I paid 79.88, with free shipping. Well worth it :)
 
JesterJJZ said:
This is what I want right?

http://www.ebay.com/itm/SLAEG-Intel...Processors&hash=item1e6f86dbc7#ht_2284wt_1154

Seems like a good deal no?

Going from Original Dual Core 2.66, worth it?
Click to expand...

Hi All, ( first time poster )

Just a quick question regarding the purchase of Xeon X5355 chips from the link mentioned above in post #358 by JesterJJZ ... How are you certain to get a "matched pair"?

I understood the success of this upgrade to a Mac Pro 1.1 was that the processors were of the same part number ( eg: SLAEG ). No?

Apparently people here have purchased from this ebay seller, did you receive a matched pair? If so, how did you ensure that you would, if not, have you still had success with the upgrade?


Many thanks,
R
 
You just want to be sure they are both the same stepping - SLAEG (X5355's)

Each processor has the step code printed directly on the lid/cap
 
michael_aos said:
I'm pretty darned tempted to give this a try.
Click to expand...

Do it. It's cheap, and easy.

If you have EVER built a PC or worked on one, or mechanically inclined even a little, you can handle it.

It goes quick.

2 most difficult things for me were getting the long allen (had to make 5 stops to find one)
And, removing the front fan. That joker was stuck!

Outside that..... gravy!
 
Just did it

I ordered two sets of matched pair X5355 cpus for $140/set, shipped. Got a tube of "Ceramique 2" thermal paste from Radio Shack for $8. Already had the long 3mm hex.

Everything went smoothly except for stripping one of the bottom two screws on the memory cage, so I had to brute-force/flex the cage to the side enough to remove the heat sink cover. Also, the grey fan assembly can be a b*tch to get out, just have to do a lot of wiggling and prying.

Did two Mac Pro 1.1 cpu upgrades. Both Geekbenched out at around 5200 prior to upgrade, now both score around 10400.

I did a real-world test of a couple things:

Opening Photoshop:
before - 22s, after - 6s

Opening 1.15gb file in PS:
before - 27s, after - 7s

FYI, on the "matched pair" thing: I found out that the only thing that needs to match is the power stepping, which is expressed in the five capital letter code, like "SLAEG". As long as that is the same for both processors, they are "matched".

One minor thing: After the installation and the updating of the firmware, on the first restart, the "about this mac" box and Geekbench indicated only one processor installed. I thought, "oh crap" as I wasn't looking forward do dismantling everything I just put back together. Another restart and it was fine. This happened on both machines I upgraded.

Temps indicated in iStat are right around 39˚-40˚ when doing stuff like browsing, e-mail
 
Last edited:
Just completed the upgrade last night... Worked like a charm. Bought a matched pair of x5355s from eBay for $140 as well. Geekbench is showing scores of 9218 & 10193. I also updated the firmware so the system reads as a 2,1 instead of a 1,1... I did it before installing the CPUs so I'm not entirely sure if it's required or not, thought I felt like after browsing the forums before the upgrade that it was... :rolleyes:

I actually had the most trouble removing the CPU heatsink cover, I was worried about breaking off a plastic tab or two. The front case fan was relatively simple to remove (just wiggle and firmly pull perpendicularly to the case/mobo), I also found that if you take the 4 screws all the way out of the memory cage you can remove the rear case fan after unplugging it's 4-pin connector and that gives you a really nice open area to work with. It also gave me the opportunity to remove every visible spec of dust.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back