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

Can 4850 iMac owners help find the cause of the freezing? Is it WiFi?

avihappy

avihappy

macrumors 6502
Original poster
Nov 15, 2006
422
1
I'm using this thread to get various data on what may be causing these iMacs to fail.

Theories Tested So Far:
  • Memory "MemVendorID" [FALSE, values same in freezing and non freezing iMacs]
  • drvincom and SG values [FALSE, it was inconsistent]
  • Manufacture Location [FALSE, all same factory]
  • Incorrect CPU fequency in hardware test [Probably Irrelevant]
  • WiFI bug theory (proposed by jebiwan) [TESTING, BUT LOOKING VERY GOOD]

Could all iMac owners (including owners of those that work fine) do this test, then fill out this questionaire (the link is below)?

TEST (broken iMac owners):
  • If you are using your broken iMac with WiFi on, do this test. If your iMac is broken and you don't use WiFi, just fill out the questionaire.
  • Try to use your iMac with Airport OFF (you can use ethernet).
  • If your iMac is broken, but it works fine with the WiFi off, this this test was "successful"
  • If it still freezes, then this test has failed.

TEST (working iMac owners):
  • If you are not using your iMac with WiFi, then you should perform this test. If you are using WiFi and the iMac works, just fill out my questionaire.
  • Turn on the WiFI and connect to a network.
  • See if the iMac works or not.

Questionaire:
Go here to submit data in a form that I can use more easily: http://www.formspring.com/forms/?595815-QBzGirSBMT

Others can view the submitted data here (make sure you select to view all columns): https://www.formspring.com/admin/data.php?595815dGKYXRw1oaK

Thanks to all who contribute!
 
J

Jammers

macrumors regular
Apr 17, 2009
103
8
iMac is freezing when under load - e.g. playing a game or watching a flash video.

"ATY,MemVendorID" = <0600>
| | | | | "ATY,MemRevisionID" = <0200>
 
T

Tom Dahl

macrumors member
Apr 9, 2009
65
0
Massachusetts USA
Does your iMac have the freezing problem: (Answer YES or NO): NO [out of 8 days use so far]
What is the number for "ATY,MemVendorID": (For example, 600): 600
What is the number for "ATY,MemRevisionID": (For example, 100): 200
 
G

GamaFu

macrumors regular
Apr 12, 2007
191
0
Does your iMac have the freezing problem: YES (5 times over three days)
What is the number for "ATY,MemVendorID": 600
What is the number for "ATY,MemRevisionID": 200
 
J

jaw1034

macrumors newbie
Apr 17, 2009
28
0
Does your iMac have the freezing problem: (Answer YES or NO) YES
What is the number for "ATY,MemVendorID": (For example, 600) 600
What is the number for "ATY,MemRevisionID": (For example, 100) 200
 
O

OBSIA

macrumors newbie
Apr 18, 2009
11
0
Does your iMac have the freezing problem: (Answer YES or NO) YES (12+x times in 2days)
What is the number for "ATY,MemVendorID": (For example, 600) 600
What is the number for "ATY,MemRevisionID": (For example, 100) 200
 
avihappy

avihappy

macrumors 6502
Original poster
Nov 15, 2006
422
1
J&JPolangin said:
...you'd think with the delay for the release of this product the GPU should have been right from the start.

This has got to be frustrating = expensive :apple: beta testers...
Click to expand...

While I understand that you are frustrated, this thread's only intention is to gather some raw data on these failing iMacs. There are other threads where you can vent...

Thanks! :)
 
S

Sermon

macrumors regular
Apr 19, 2009
111
0
Does your iMac have the freezing problem: (Answer YES or NO): NO [out of 6 days use so far]
What is the number for "ATY,MemVendorID": (For example, 600): 600
What is the number for "ATY,MemRevisionID": (For example, 100): 200
 
d510r9

d510r9

macrumors newbie
Mar 3, 2009
18
0
My iMac is a 2.93GHz 4850

Does your iMac have the freezing problem: (Answer YES or NO): NO
What is the number for "ATY,MemVendorID": (For example, 600): <0600>
What is the number for "ATY,MemRevisionID": (For example, 100): <0200>
 
GonzoRob

GonzoRob

macrumors 6502
Sep 1, 2004
271
0
perhaps it isn't the memory . could people with working imacs please use the command :

ioreg -l | grep ATY

put the results on pastebin.com and let us have a look ... it may be something else (my crashing imac is http://pastebin.com/m6eb9fc77)

Also, it could also be the case that people with 'working' imacs simply haven't stress tested their GPU's hard enough as my mac is pretty stable when not under load.

If you can run this app (http://www.daionet.gr.jp/~masa/rthdribl/) in windows boot camp for an hour (demo mode, full res, mull multisampling) then you're ok. If it freezes, you've got the same problem as the rest of us. Personally, I think it should be the standard test :)
 
GonzoRob

GonzoRob

macrumors 6502
Sep 1, 2004
271
0
checked Sermon pastebin,

The only two differences I can spot are :

"ATY,SG" =
my value : <08f01a47>
your value : <08a0f646>

and

"ATY,drvincom" =

my value : <8da6620000700347>
your value : <8da662000050ff46>


However I have a feeling that these values will be different the whole time :confused:
 
d510r9

d510r9

macrumors newbie
Mar 3, 2009
18
0
GonzoRob said:
perhaps it isn't the memory . could people with working imacs please use the command :

ioreg -l | grep ATY

put the results on pastebin.com and let us have a look ... it may be something else (my crashing imac is http://pastebin.com/m6eb9fc77)

Also, it could also be the case that people with 'working' imacs simply haven't stress tested their GPU's hard enough as my mac is pretty stable when not under load.

If you can run this app (http://www.daionet.gr.jp/~masa/rthdribl/) in windows boot camp for an hour (demo mode, full res, mull multisampling) then you're ok. If it freezes, you've got the same problem as the rest of us. Personally, I think it should be the standard test :)
Click to expand...


I ran my 4850 in Windows 7 playing Crysis for a 5 hour straight session...high settings, 2xAA, dx10...not a single freeze.
 
S

Sermon

macrumors regular
Apr 19, 2009
111
0
GonzoRob said:
checked Sermon pastebin,

The only two differences I can spot are :

"ATY,SG" =
my value : <08f01a47>
your value : <08a0f646>

and

"ATY,drvincom" =

my value : <8da6620000700347>
your value : <8da662000050ff46>


However I have a feeling that these values will be different the whole time :confused:
Click to expand...

Looks like it - jus checked another one and these values are different to ours.
 
N

nss.aries

macrumors member
Feb 16, 2009
56
0
Does your iMac have the freezing problem: YES
What is the number for "ATY,MemVendorID": 600
What is the number for "ATY,MemRevisionID": 200
 
avihappy

avihappy

macrumors 6502
Original poster
Nov 15, 2006
422
1
This theory was too inconsistent, and I think it has been falsified.

Here is some potentially interesting data I have so far:

Based on the 3 pastebin data sets I have so far:
Both the "ATY,SG" and "ATY,drvincom" variables on the 1 working iMac end in the number 6, while the data from the two broken iMacs has these variables end in 7.

WORKING IMAC
"ATY,SG" = <08f0f746>
"ATY,drvincom" = <8de6610000e0fd46>
Click to expand...

EDIT, Adding new entry:
SERMON'S WORKING IMAC
"ATY,SG" = < 08a0f646>
"ATY,drvincom" = < 8da662000050ff46>
Click to expand...

BROKEN IMAC 1
"ATY,SG" = <08402047>
"ATY,drvincom" = <8de6680000d00a47>
Click to expand...

BROKEN IMAC 2
"ATY,SG" = <08f01a47>
"ATY,drvincom" = <8da6620000700347>
Click to expand...

This is the only noticable patten I can see with my tiny dataset, and it's not enough data to make a conclusion. If more people can contribute their data, we might catch on to a proper pattern!

EDIT: Just looked at Sermon's Pastebin at http://pastebin.com/f5bb5a720 , his working iMac holds up to this patten as both of these numbers on his iMac end in 6! My hope is getting better and better on these variables!
Click to expand...
 
H

Hoya1

macrumors newbie
Mar 27, 2009
28
0
My iMac is working, too, and also seems to fit the pattern, though my ATY,SG and ATY,drvincom both end not in "6" but rather in "6b"
 
avihappy

avihappy

macrumors 6502
Original poster
Nov 15, 2006
422
1
OBSIA's data makes me a little worried, but still want to see more data before trashing the theory.

I have also realized that the "ATY,SG" variable is not very reliable, as I have seen multiple "working" iMacs that have it end in 6 (contrary to the theory).

So I think that "ATY,drvincom" will be the variable I would pay more attention to. It has been rather reliable so far (the exeption being OBSIA, whose broken iMac ended had this end in 6).

EDIT: OK, based on OBSIA and iMac4850User's data, I think I am ready to throw out my theory. Oh well. I think maybe now we can check for mfg location?
 
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