Firmware update and SATA II hard drive

Before applying the new Performance Update 1.0, I had rolled back to EFI 1.6. Even with 1.6, if I tried anything I/O intensive such as running a virtual machine through VirtualBox for Windows Vista to run Adobe CS3, it would randomly freeze. With EFI 1.7 it was always freezing, or as some of you would refer to it as "beachballing", so I stayed with EFI 1.6 until now. Yesterday I first applied the Performance Update 1.0 with EFI 1.6, and then went back to EFI 1.7 and everything is working great so far! I have not gotten any freezes anymore thus far, and have been able to run VirtualBox and Photoshop smoothly, so whatever they updated in the performance update seemed to have helped in my case. *knock on wood*. By the way, I never had issues when running Bootcamp, w/ Windows XP or Vista. Hopefully Apple will release a SL CD with this update already included. I hope others are able to see resolution to their issues as well, best of luck everyone!

I had a decently working MacBook Pro 2.53 Ghz with the stock 250 GB HDD. I had to apply the hdapm for it to be completely BB free. I felt that the stock HDD was a little slow and thus replaced it with a Seagate 7200.4 500 GB (non G)!

Thats when my nightmares started. HDAPM didn't help one bit, and I was left scrambling.

I stumbled upon a suggestion by someone to downgrade to EFI 1.6! Magic bullet. In EFI 1.7, the comp was literally unusable. Beachballs every 2-3 mins lasting 30 seconds... even the **** OS reinstall took 2-3 hours (I feel asleep waiting for it.)

However, I install Performance Update 1.0 and went back up to EFI 1.7, and everything seems fine 🙂 Had 2 BBs till now (very isolated and lasting around 15 seconds) but otherwise a-ok 🙂

So, the trick for those in 1.7 is to jump back to 1.6, apply the Performance Update, and then reupdate to 1.7 ?

Oh Goooshh ! I'm not sure I dare to try this.

Saharis wrote:
So, the trick for those in 1.7 is to jump back to 1.6, apply the Performance Update, and then reupdate to 1.7 ?

Oh Goooshh ! I'm not sure I dare to try this.

Shouldn't make a different if you apply the update with 1.6 or 1.7, and I wouldn't suggest flashing EFI needlessly... I applied this on top of 1.7 and no longer have problems, although my issues were never as chronic as some (no CRC errors accumulated throughout the entire ordeal)

Looks like the Performance update with its new kernel extension is mitigating the SATA issue in some way.

I really would be interested to see a Harddrive Benchmark of somebody who does not see the beachballs anymore. I wonder if the new driver is just reducing the SATA speeds to make this go away for OSX installs.

For comparison, I just did a test with the AJA System Test tool and my SSD performance is 147.8MB/s write and 191.2MB/s read on my MBP15 (which did not show the EFI 1.7 issues anymore after the logic board swap).

Ollie.

Ollie.

I just ran the AJA disk read/write test, 128mb and my results were: Write 87MB/s, Read 229 MB/s.

I don't know what these numbers show.

13" MBP (MB991LL/A)
X25-M (SSDSA2MH080G1GN)

Mac OS X 10.6.1 with 64-bit kernel
EFI 1.7 firmware (MBP55.00AC.B03)
SSD 045C8820 firmware
SMS deactivated
Performance Update 1.0

Note the 16GB test run for 16Kib, 32Kib, 64KiB, 128KiB, 256KiB, 512KiB, 1MiB, 2MiB, 4MiB, 6MiB and 8MiB blocks:

AJA System Test results

awerty9999 wrote:

Wan Chai Man wrote:
I did this and trebled the AJA test figures on my 13" MBP:

1) Did a clean install of Snow Leopard

2) Installed the performance update

Dunno which one (or both) was the solution but I am not complaining :):)

It is not the réinstall, I can assure you of that, I did it myself and it didn't work.
Waiting to see if the performance update changed anything.

Well, you can rest assured that I am a very happy camper:)

And on 22nd October I will be an even happy boot camper!

For all those using AJA System Test: Try running a test (512MB is what I use) while a video file (of a few hundred MBs in size) is playing in Quicktime. Run it maybe 5-10 times and see if it freezes or not.

Especially for those who say they still have had beachballs (but say it's fine anyway) like KiD0M4N and somebody else whose post I can't find right now. Or others who have seen the symptom of incrementing CRC errors under EFI 1.7 (and not under EFI 1.6).

It's been hypothesized before, but perhaps a side-effect of the kernel extension patch is that it makes freezing from the firmware interface issue less noticeable under normal use. I use the video playback + AJA System Test method because it has proven quite reliable in bringing out the issue (another method is playing a video, then making a copy of a large file), but it cannot be considered "normal use".

Oliver F wrote:
Looks like the Performance update with its new kernel extension is mitigating the SATA issue in some way.

I really would be interested to see a Harddrive Benchmark of somebody who does not see the beachballs anymore. I wonder if the new driver is just reducing the SATA speeds to make this go away for OSX installs.

For comparison, I just did a test with the AJA System Test tool and my SSD performance is 147.8MB/s write and 191.2MB/s read on my MBP15 (which did not show the EFI 1.7 issues anymore after the logic board swap).

Ollie.

Ollie.

You also had your SATA cable replaced, right?

I'm curious if the replacement looks different than the original one. (some place with tear-down pics, like iFixIt, could probably refresh your memory if you don't recall the original)

I replaced my SATA cable and it looks the same with the exception of some printing on it. Frankly I don't know how you'd tell them apart.

Guys/Gals:

Well I finally got my 15 inch Mid-2009 MB Pro EFI 1.7 Firmware Update issues resolved.

Owners of these Mac Laptop units, who have had 3rd Party SATA // 7200 RPM HDs installed, had issues with the computer not restarting or having constant "beach balling" issues after updating the Firmware to EFI 1.7 from the previous EFI 1.6 version which came installed with the laptop. There are 116 pages of discussion the Apple Forum involving these issues.

Apple finally released a "roll back" of EFI Firmware 1.7 to EFI 1.6 Firmware in order to allow the use of these 3rd party faster HDs--but at less than full performance. I took my MP Pro to the Apple Store and the "Apple Genius" rolled the Firmware back to EFI 1.6. That allowed me to have the larger 500 GB Seagate purchased HD re-installed by my Apple dealer--QuadStar--who temporarily got me up and running by initially swapping out my newly purchased 500 GB Seagate HD with the original factory supplied 250 GB Hitachi HD.

I was limping around OK with this modification but I could not use my 500 GB SATA // HD to its fullest capacity.

Just last week Apple issued a *Performance Update 1.0* package to resolves this issue.

I installed the Performance Update package--and after installation could see no difference as I was still on the "Roll Back" to EF 1.6 Firmware.

Today I called the Apple site and confirmed my reinstall of EFI 1.7 to get full capability of the larger faster HD.

The Apple Care rep indicated that I was "Good to Go" and I should re-install the EFI 1.7 Firmware Update. I was very cautious as this locked up my computer twice before.

Now I am happy to report, the re-install of EFI 1.7 did not lock up the computer and so far every thing is finally performing smoothly for now--after since about the end of July when all of this grief stared after the installing the EFI 1.7 Firmware update initially.

Here is my HW Info:

Model Name: MacBook Pro
Model Identifier: MacBookPro5,4
Processor Name: Intel Core 2 Duo
Processor Speed: 2.53 GHz
Number Of Processors: 1
Total Number Of Cores: 2
L2 Cache: 3 MB
Memory: 4 GB
Bus Speed: 1.07 GHz
Boot ROM Version: MBP53.00AC.B02
SMC Version (system): 1.49f2
Serial Number (system): XXXXXXX XXXXX
Hardware UUID: 961058BB-22A8-5F4C-A2DE-386511CF01C1
Sudden Motion Sensor:
State: Enabled

Here is my Serial ATA info:

ST9500420ASG:

Capacity: 465.76 GB
Model: ST9500420ASG
Revision: 0002SDM1
Serial Number: 5VJ092LJ
Native Command Queuing: Yes
Queue Depth: 32
Removable Media: No
Detachable Drive: No
BSD Name: disk0
Mac OS 9 Drivers: No
Partition Map Type: GPT (GUID Partition Table)
S.M.A.R.T. status: Verified
Volumes:
RJG MBP:
Capacity: 465.44 GB
Available: 233.6 GB
Writable: Yes
File System: Journaled HFS+
BSD Name: disk0s2
Mount Point: /



Now I am a "happy camper" finally!

bobg

Robert... It's great that you are no longer having issues with the 1.7 firmware. However... just to elaborate... no notebook hard drive comes close to saturating the bandwidth of SATA 1.5GB. Even current high end SSDs just occasionally do so. The 1.5GB issues isn't likely to ever be an issue with HDD as they have been at their current level of performance for a number of years now. However, we can expect SSDs to get progressively faster (as they have done in the short time they have been available). Hopefully we'll be able to take advantage of these improvements in performance improvements as they roll them out.

Robert Gulyas wrote:
Guys/Gals:

Owners of these Mac Laptop units, who have had 3rd Party SATA // 7200 RPM HDs installed, had issues with the computer not restarting or having constant "beach balling" issues after updating the Firmware to EFI 1.7 from the previous EFI 1.6 version which came installed with the laptop. There are 116 pages of discussion the Apple Forum involving these issues.

Apple finally released a "roll back" of EFI Firmware 1.7 to EFI 1.6 Firmware in order to allow the use of these 3rd party faster HDs--but at less than full performance.

I was limping around OK with this modification but I could not use my 500 GB SATA // HD to its fullest capacity.

Just last week Apple issued a *Performance Update 1.0* package to resolves this issue.

Robert, I'm glad the issue you had appears to be gone, but JoeyR is right... if you were "limping around" under EFI 1.6 with a traditional hard drive, it wasn't due to the SATA I limitation.

Also, to be clear, Apple has never officially said that Performance Update 1.0 was intended to resolve any EFI 1.7 related issues. In fact, all signs point to the exact opposite: it's a OS patch (while the EFI 1.7 problems have been OS-independent), the list of affected machines consists of 15 models ranging from 2009 iMacs and Mac Minis to 2008 MacBook Airs (while EFI 1.7 is strictly limited to mid-2009 13" and 15" MacBook Pros), etc.

fishbert:
I was not able to see any physical difference in regards to the new SATA cable. I did see it being swapped, but was a little far far away. from what I can tell looking at it now comparing to the pictures on the web, it looks pretty much the same.

I still believe there is a chance that the culprit is the EFI upgrade process itself. Why I say that is:

1) The issue is hit and miss. We had working and non-working SATA MBP all along. There does not seem to be a relation between the issue and the build date. I do not believe that there is a HW rev difference between the working/not working MBP looking at the history at hand. The thread is long, but not as long as too suggest that its an issue affecting the whole customer base.

2) During the upgrade process the FW is copied to the EFI partition from where it is executed during boot. Hence there is a chance, that any possible issue addressed by EFI 1.7 or any issues during writing the FW update from the download to the EFI partition could prevent a "clean" FW burn (which would have to transfer over the SATA bus). I looked at the binaries doing the upgrade and i do not see any specific CRC checks implemented in that software. When I installed mine the second time, I took extra care and done it from a fresh install.

3) Some users have now reported, that after installing the Performance update and re-upgrading to EFI 1.7 from 1.6 their issues are gone while seemingly having full SATA II performance as demonstrated by the Benchmarks posted. This potentially could mean that FW burn this time happened correctly, rather than the issue being fixed by the "Performance upgrade" which best case would only smooth over the symptoms rather than fixing them.

There really is no harm in trying the upgrade. The 1.6 downgrade is available to us as a worst case measure (or preferably going to the Apple store for the downgrade if you are near one).

I for one did not have to resort to this as my issue was fixed after the second upgrade.

The only caveat is still that my issue was resolved by the logic board and cable swap early on (before the downgrade utility was available) but given point 1) this seems to be not the case.

I really wonder if people having the issue resolved would see issues under Bootcamp.

Ollie.