Weeks have passed since Apple's announcement of the Mac Pro, and while we wanted to conclude our look at the Mac Pro much earlier, like many Mac Pro users we ran into some serious performance issues under Windows XP.

With the Mac Pro performance issues resolved and some more time with the system under our belts, we're able to bring you the final part in our Mac Pro coverage. This time we're focusing on upgrading the memory and CPUs in the Mac Pro, as well as looking at its performance as a PC running Windows XP.

As a high end Xeon based machine that can run both Mac OS X and Windows XP, the Mac Pro has the potential to be the power user's dream. Today our task is to find out just how upgradable this machine is and how well it runs XP, if it can truly be your only system if you're a Mac and PC user.

FBD Revisited

Thus far the only real downside we've seen to the Mac Pro is its use of Fully Buffered DIMM (FBD). As we mentioned in our initial article discussing the Mac Pro's specifications, the FBD spec calls for a serial interface between memory controller and memory modules, while allowing the chips on the memory modules themselves to be regular mainstream DDR2 devices. A FBD memory controller talks to an AMB (Advanced Memory Buffer) on each memory module, which acts as a translation hub and buffer for all communication between the DDR2 devices on the module and the requests from the memory controller.

The major benefit of FBD is the ability to feature more memory modules per channel (up to 8 per channel), offering greater capacity for high end servers and workstations than even registered DDR2. The downside to FBD is that there is significant overhead and latency introduced by using a packetized interface and using the AMBs to translate from one interface technology to another (FBD to DDR2).

As we mentioned and proved in our previous articles, the number and configuration of FB-DIMMs in your Mac Pro can affect performance. The Intel 5000X chipset in the Mac Pro features two 144-bit FBD branches, each being the width of two FB-DIMMs (effectively giving the chipset four "channels"). Therefore you need at least two FB-DIMMs in the system (the width of a single FBD branch), but ideally you'd need at least four to have a hope of attaining peak bandwidth.

As some of our readers (and Intel) pointed out, the story doesn't just end at needing four FB-DIMMs. The rank of the FB-DIMMs can impact performance as well, and ideally each of your FB-DIMM modules would be dual rank modules. The rank of a DIMM is determined by dividing the width of all of the devices on the module by the width of the module itself. For example, a single rank FB-DIMM would have 9 DDR2 devices each being 8-bits wide. A dual rank FB-DIMM would be composed of 18 DDR2 devices, each still being 8-bits wide. All of our 512MB FB-DIMMs are single rank modules, while our 1GB and 2GB modules are dual rank.

The story doesn't end with rank though. Because of the dedicated read and write lanes between the memory controller and the AMBs on FB-DIMMs, you can be reading from one FB-DIMM while writing to another. So in theory, if you're running an application (or combination of applications) that have a lot of concurrent reads and writes going on you could stand to benefit from having more than one FB-DIMM per channel.

Based on all of the above information, it would seem like your best bet is to stick as many dual rank FB-DIMMs as you can afford in your system, and if that were the case then we'd be able to move on from here. Unfortunately it's not, because as we mentioned in previous articles, the more FB-DIMMs you have in your system, the higher access latencies will be to those additional FB-DIMMs.

What we then end up with is a tradeoff between more bandwidth and higher latency, so which route do you take? We've done a lot of testing and most of our tests seem to favor the four dual-rank FB-DIMM module configuration, but the number/configuration of modules really depends on your particular needs. We're still testing to figure out what the tangible real world performance differences are between the multitude of memory configurations, but for now just know that if you need maximum bandwidth you'll want 8 dual rank FB-DIMMs, but if you want lower latency you'll want fewer modules. Whether or not you'll see a performance difference will depend mostly on the application(s) you're running.

Third Party Memory Modules
Comments Locked

72 Comments

View All Comments

  • bobsmith1492 - Tuesday, September 12, 2006 - link

    Well, thank you for the enlightenment. I will never use Vista now that I know I cannot use more than 64GB of RAM.
  • blwest - Tuesday, September 12, 2006 - link

    Who did that comment come from, something about never needing more than 64K? Was it Bill Gates?

    I already have customers that use Linux over MS because they utilize 64GB of ram TODAY. Even when Vista is released, they will still be using Linux.
  • bobsmith1492 - Tuesday, September 12, 2006 - link

    Well, I've doubled the amount of ram in my _personal_ computer twice in about 5 years along with one upgrade of Windows (98-XP)... so, in 5 more, I'll have 8GB, 10 more: 32, and in about 12.5 years, I'll be up to 64GB, by extrapolation. :P Anyway, I don't expect Vista to last that long, so I guess I'm ok (heck, I'll use XP for a few more years anyway, no doubt - XP was out for a few years before I upgraded).

    As far as the 64GB of RAM, if you need that for your application, you probably wouldn't even dream of using a closed-source OS for it anyway, so I think the RAM limitation isn't really that relevant (maybe for servers...)
  • greylica - Tuesday, September 12, 2006 - link

    Course the common user will not have to use that quantity of Ram, as far as I know, they are still using 256MB at nmaximun in order to run only apps for office/small production.
    Here in Brazil the most of them are using at maximun 512 MB of Ram, but when we think in a machine different, like a MAc, or then a graphic Workstation, or then an data seismic or a test evaluation machine, 64 GB in 5 years will be a cap, a limit.
    Anyway Linux can save our souls to this switch Microsoft is trying to do again, and again...

    A simpler crowd simulation in XSI can achieve 16 GB of Ram in a workstation...

    Vista Premium can accept more than 4GB, course, but in the 32 bit usage, the same old 3GB switch is aplicable, but I don´t even imagine why they at the same tima reserve 4GB for the kernel, and limit Vista to 1/2 of the potential.
    It´s a same old problem coming back again.

    Even 2000 accepts PAE, but doesn´t serve for nothing, course they can give a patch, but, do they want ???

    If a patch for ths issue is released, this topic can be sent to oblivion.

    Microsoft, give us a patch to 3GB switch right now !!!!!!
  • mostlyprudent - Tuesday, September 12, 2006 - link

    I have to agree with the first part of your post. I read the article and thought.."Why?". I have always understood the appeal of a Mac to be that you could upack it and get to work. If we start talking about running other OSs on a Mac or hacking software/drivers for better hardware supprt, then why pay the premium for a Mac?
  • lopri - Tuesday, September 12, 2006 - link

    Well, sort of.. Remember that now Mac is a PC with the ability to dual-boot between XP and OS X. Other than the anatomy description for CPU upgrading, I don't see anything Mac-specific about this article. You can replace the tested Mac with an equivalent workstation from Dell, HP, etc. and the end result would be the same. It's also very questionable how much appeal this article would have to an 'average' apple user - who wants a computer that just works, without worrying about upgrading or fixing, and prefers to use for creative works or entertainment. (This isn't my opinion, btw. I happened to read a few articles on Apple/Mac @DailyTech, and it's how Mac users described themselves) I totally understand this article is a sequel/finale to a previous article, but I think some people would agree with me. (If not, oh well.. :) )

    From this viewpoint, I'd like to make a comment that Anand's obsession on Apple (be it hardware or software) has been quite over-the-top already. Even though this article isn't really about something unique to Apple, he doesn't skip a single product that Apple produces. (I still can't believe he "reviewed" a mouse just because it's from Apple. There are many more superior and innovative mice in the market, you know.)

    Whether it's his personal preference towards Mac/OS X or a business strategy to expand the readership of this site isn't clear to me. But I can't help but notice the imbalance and ask why.
  • Anand Lal Shimpi - Tuesday, September 12, 2006 - link

    The point of the article was for users that want to run OS X as well as Windows XP, rounding out the performance comparison by showing what sort of XP performance you can expect out of the Mac Pro.

    As for our Mac coverage, we definitely don't review every Apple product that comes out, but the ones that we do focus on are those that are most interesting to the community. The vast majority of our content is still PC focused, but whenever there's a big Apple release we will do our best to cover it just as we do major releases on the PC side.

    Take care,
    Anand
  • JAS - Thursday, September 14, 2006 - link

    PC = Personal Computer, regardless of whether it is running OS X or Microsoft Windows.

    Please continue with the Macintosh coverage, Anand. I appreciate the excellent work that you do, and even bought a Mac because of one of your reviews. What a fabulous computer.
  • Calin - Tuesday, September 12, 2006 - link

    :)
    I'm sure Anand wants to expand the readership of his site to the 5% Mac users :)

    (no offence, Anand) but I think this was more like a toy, that is getting better and better. "I bet I could use only this Mac for everything" could have been the idea of the first article, and in the end it seems the Macs were even better than that (whether by look and feel, ergonomy in user interface, invulnerability to almost all viruses/worms/... (the first Mac used by Anand was Power based), other reasons or a combination of all the above).

    On another note, I wonder how well those new Macs will survive in the virus world, now that they now have the x86 processors
  • JAS - Thursday, September 14, 2006 - link

    quote:

    I wonder how well those new Macs will survive in the virus world, now that they now have the x86 processors.

    If you mean running Microsoft Windows on a Macintosh, then it is as susceptible to spyware and viruses as any other MS Windows computer. These problems are tied to the operating system, not the microprocessor. OS X is as secure running on Intel as it is on PowerPC chips.

Log in

Don't have an account? Sign up now