Even if I'm still using this 2009 Mac Pro in 2019... I won't be going back to anything shipped by Microsoft... ever! Cue the song by Taylor Swift "We are never ever getting back together" 
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Pros on the "workstation fence", what will it take to get you into a new MacPro?
- Thread starter echoout
- Start date
- Sort by reaction score
I would wait and see what the 2013 Mac Pro has to offer. I am not hoping to wish for the fastest top of the line model since my budget cannot afford the price level anyway. I usually go for the midrange model. So far satisfied and happy with the current Mac Pro I am using as it gets the job done with my work.
For the new Mac Pro, I hope it would have USB 3.0, faster PCEi, better GPU and CPU choices. Thunderbolt would be great but in my case not yet an urgent need. I do heavy graphics work and some medium After Effects projects. Twice I switched to Windows PC but ended up going back to Macs. It was more of the user friendliness and stability of OSX. Not sure if I'll go back to windows.
For the new Mac Pro, I hope it would have USB 3.0, faster PCEi, better GPU and CPU choices. Thunderbolt would be great but in my case not yet an urgent need. I do heavy graphics work and some medium After Effects projects. Twice I switched to Windows PC but ended up going back to Macs. It was more of the user friendliness and stability of OSX. Not sure if I'll go back to windows.
If there is indeed a new MacPro, I see the same connections as their laptops as a no-brainer as far as TB (I think) and USB 3.0 and a speed bump as minimum. RAM, GPUs and I/O cards seem as plentiful as ever after the fact so that's no big deal.
So I guess I'm really watching PCIe slot speed and count, and processor core speed and count the more I look at the situation. More internal storage and the fabled 2TB SSD would be nice but I can live without them.
So I guess I'm really watching PCIe slot speed and count, and processor core speed and count the more I look at the situation. More internal storage and the fabled 2TB SSD would be nice but I can live without them.
I think the title of the tread "Pros on the "workstation fence", what will it take to get you into a new MacPro?" is misleading because the definition of a Pro includes knowing about their systems specifications and reasonable expectations. Thus no "Pro" will ever be on the fence - they will know what they need. That's part of being a Pro after-all.
And of course you don't see Apple offering any specific specs for their 2013 systems... Because they haven't published a spec yet.
After they do then you can determine if it's fast enough for what you need or how it stacks up against offerings from other vendors.
Last edited:
Perhaps for different reasons, I'm in the same boat and have considered the same/similar replacements if Apple doesn't come through. Basically, I'll accept the usual restrictions that come with any Apple product in terms of flexibility in exchange for reliability/support/etc. However, that assumes I can get clock speed (which I need for certain apps as much, or more than, cores alone) via a modern CPU/Mobo architecture, and a HOT GPU (i.e., Titan). Round it out with USB 3.0, sufficient PCI-E slots, and plenty of Wattage, preferably with a few extra connectors.
I already have multiple computers at home and in the office, so the obvious option is, buy both! But, that doesn't sit well with me. It isn't the money so much as I still like the idea of having a "main" or "go to" box for the heavy stuff. Assuming the premise of whether to totally jump ship is based upon the new Mac Pro being either vaporware or some silly modular or otherwise iOSified contraption with limited speed/ports/expansion (not a TB fan BTW), for me, it purely becomes a decision of which Windows route to go. OS-wise Win 8 is out...period. Win 7 with an XP VM, may it live forever.
Then it becomes an issue of Xeon v. i7 Extreme CPU choices. Boxx may offer an OC'd Xeon and have a great rep for service/support of a sturdy machine, but their pricing across the board is absolutely laughable and makes Apple look dirt cheap. Origin (ex Alienware inventors) will build you a similar maxed out Xeon workstation or overclocked i7 past 5 GHz for less money and the same, if not greater, reliability, service, and support. If mainstream off the shelf is good enough, you will not beat the HP Z820. If their post-sale tech support is remotely close to their pre-sales tech support, it's almost Apple-like. Their hardware is rock solid too.
The consensus across all of these threads seems to be that (for the most part) we know what we want in a new machine and are crossing our fingers that Apple delivers (soon). It's the waiting game for those that are hanging on with hope, and the switchers conundrum for those that would like to spend more time getting work done rather than working within the limitations of their existing solutions.
Personally, I'd love to deliver more for my clients in a shorter amount of time, but there's only so much you can do when deadlines need to be hit and you still haven't hit the render or export button. But it's been this way FOREVER.
Prior to Mac Pro workstations we're using today, many of us were on expensive SGI boxes, custom-rigged Avid machines, or custom hardware without an OS-driven environment (unless you were a coder). They weren't very easy to use, took up lots of space, and we pushed the limits of what they could do. The difference is, these machines weren't (realistically) available to the masses and you needed to be at a facility that invested in them. That market has changed drastically.
I'm fine with 3rd party upgrades and customizations - don't need it all to be BTO. There's no chance Apple is going to sell me all of the hardware I need, plus I already own and have access to almost all of it. In the end, I wish we'd have more transparency and not need to rely on rumors to make a roadmap, but those are too high expectations from Apple in regards to the pro community it seems...
Personally, I'd love to deliver more for my clients in a shorter amount of time, but there's only so much you can do when deadlines need to be hit and you still haven't hit the render or export button. But it's been this way FOREVER.
Prior to Mac Pro workstations we're using today, many of us were on expensive SGI boxes, custom-rigged Avid machines, or custom hardware without an OS-driven environment (unless you were a coder). They weren't very easy to use, took up lots of space, and we pushed the limits of what they could do. The difference is, these machines weren't (realistically) available to the masses and you needed to be at a facility that invested in them. That market has changed drastically.
I'm fine with 3rd party upgrades and customizations - don't need it all to be BTO. There's no chance Apple is going to sell me all of the hardware I need, plus I already own and have access to almost all of it. In the end, I wish we'd have more transparency and not need to rely on rumors to make a roadmap, but those are too high expectations from Apple in regards to the pro community it seems...
I think the title of the tread "Pros on the "workstation fence", what will it take to get you into a new MacPro?" is misleading because the definition of a Pro includes knowing about their systems specifications and reasonable expectations. Thus no "Pro" will ever be on the fence - they will know what they need. That's part of being a Pro after-all.
And of course you don't see Apple offering any specific specs for their 2013 systems... Because they haven't published a spec yet.
Not arguing but I don't understand how you say no pro can be on the fence. I know a legion of creatives on the fence. We know what we want from the current crop, but we're speculating about the next crop. As it stands, Windows workstations are much faster and have more options. I was just trying to poll everyone, who is on the verge of a new workstation, what the minimum would be from a currently-nonexistent MacPro for you to choose that over the other options.
Nothing more, nothing less.
And you have great input. I appreciate it.
Yeah, this part of the equation is pretty wild aye. Wild meaning messed up!
----------
legion of creatives ≠ legion of professionals.
Waiting for Apple ≠ Being on the fence.
Probably just a difference in our interpretation of the phrases.
I guess most people choose Apple products because of OS X and whatever investment they have represented in the ownership of various applications and/or 3rd party hardware (cards). Besides the style of the boxes I can't think of anything else at least. The MacPro has always been a good solid choice for entry to mid-level workstation users hardware-wise (and with few exceptions, always 6 to 8 months behind). I assume whatever is released next will fall into the same category as well.
Last edited:
Yeah, this part of the equation is pretty wild aye. Wild meaning messed up!
----------
legion of creatives ≠ legion of professionals.
Waiting for Apple ≠ Being on the fence.
Probably just a difference in our interpretation of the phrases.
I don't usually refer to non-professional folks in the creative field as "creatives" but okay. And I guess I assumed that if Apple are riding Intel's roadmap (as they have sometimes done) then other workstations would also be in for changes. Speculating about new Macs is already too much, not going there with HP, Boxx, Dell or whoever.
That's actually a good question, in the past, have the PC brands moved to a new chipset/processor/etc. at the same time as Apple? I know Apple have gotten some models early (I think).
I edited my post. But yeah, typically it seems Apple is usually just shy of a year late to the table. Part of that is due to Apple's target market I think - I guess. It makes really good economical sense if their target is in the low or mid ranges somewhere that they don't design and develop for the high-end. And it's often the case that the low and mid ranges don't need the newest tech - which is often only available at first for the high end.
Other manufacturers typically follow suit. At least I can't think of any exceptions where a vendor is beating the curve.
I dunno Boxx these days but a few years ago they were targeting the same market segments as Apple's MacPro and their systems were not remarkably different. HP has a huge range and Dell is currently undergoing some stresses so I dunno where they're at right now. I guess check their respective websites to know more.
Looking at the Boxx site for the first time in a long while it seems the same. Slightly newer tech but you're paying for it!
Boxx 8980 Xtreme
- ES-2687W 3.286GHz dual procs (water-cooling [and over clocked to 4GHz?]) 16-cores
- NVIDIA Quadro K2000 2GB (can only fit 2 GPU cards in the 8980 Extreme model - 4 in the 8950 but no over clocking I think, thus not "Xtreme".)
- 8x4GB DDR3-1600 ECC RDIMM(32GB)
- One 500GB HDD
- Win8 64bit
- KB+Mouse
- 1 year support
- Intel Xeon E5-2687W Performance Enhanced Processors Liquid Cooled
- Intel® 602 Chipset
- RDIMM ECC DDR3-1600Mhz
- 8 DIMMS – OC 1980 Mhz
- 16 DIMMS – OC 1700 Mhz
. . (16) DIMM Slots - (8) 3.0 GB Serial ATA ports
. . (RAID 0, 1, and 10) - (2) 6.0 GB Serial ATA ports
- (2) Gigabit Ethernet ports
- HD Audio
. . Front Panel Mic + audio out - (6) USB 2.0 Ports:
- (2) Front, (4) Rear
- (4) USB 3.0 Ports:
- (2) Front, (2) Rear
- 1250W Power Supply
- 4U: 6.825"w x 17.815"h x 24.65"d
- Expansion options:
- (8) 3.5" hard drive bays
- (1) 5.25" drive bays
- (2) PCI Express v3.0 x16
- (4) PCI Express v?.? x8 - (I assume also v3.0)
Apple's 12-core with 3.06 procs, 32GB, 1TB HDD, and the HD 5770 1GB, is $6,949.00 with free shipping and 0% loan. But of course Apple wants about $900 for that RAM so -$750 to go with the smallest amount of RAM available and buy your own for $300.
Last edited:
Yeah, the build I did at BOXX was $11,676. My equivalent maxed out MP speed with OWC RAM and SSD, and NewEgg GTX 680 is $7307. Soooo, yeah, about $4370 for the (much) faster processors. I can make that difference up fairly quickly with my ongoing jobs, so it's there if I want to go that route.
And shipping is free from Boxx since I pass them on the way to school a few times a week.
And shipping is free from Boxx since I pass them on the way to school a few times a week.
The other workstations already made changes. That all did relatively nothing in 2011 and transitioned to new base motherboards in 2012. They will likely just keep those in 2013 with some relatively minor upgrades to v2 E5 parts (with some minor firmware upgrades) as they become available in volume this year.
There is no huge shift coming this year. It is a "tick" (process shrink) year where nothing much changes the bigger iron market in terms of ports, slots, and I/O.
Apple is way, way, way, off the reservation as far as closely following Intel's product roadmap. That may be just as well since isn't on track to deliver yearly updates and frankly at this point that is primarily what Apple needs (more predictability).
An apples to orange comparison because more PC brands have at least 3 to 4 times as many products as Apple does. There is always some churn somewhere in their line-ups because they are so massively complicated and overlapping.
That was an exceptional one time fluke which folks keep trying to turn into some kind of common and/or expected occurrence. There is very minute substance to that position.
Minus Apple and Intel working out some custom fab relationship on x86 SoC, Apple isn't getting anything in front of anyone else who puts their money on a purchase order. Custom fab work wouldn't be all the unique to Apple if Intel expands that business though. But "early access" is relatively normal for larger customers. They all have manufacturing production ramp ups and early evaluation units to roll out to their high interaction customers.
Actually they've done it thrice with Intel chips already. And if you consider the PPC line they did it with every single release.
So yes, they correctly have a reputation of minor exclusivity when it comes to core components. Thunderbolt is the most recent example I believe... Wasn't Apple the first build to come to market with a thunderbolt port? I'm actually not sure on this one but I believe so.----------
And shipping is free from Boxx since I pass them on the way to school a few times a week.
Hehehe, kewl!
Approximately $4.4K could be a dedicated batch job rendering box. Sure there are some that need everything all inside of a single box with slots but with large 4-6K differentials on pricing using more than one system works out just about as often as it doesn't. Especially when dealing with groups of 3-5, all doing similar computations on a commonly used set of data.
That's actually a good point. (Hey, you got one...)
Sometimes it's better to be a Beowulf than a Godzilla!
)Sometimes it's better to be a Beowulf than a Godzilla!
Well, if you buy the RAM and drives from someone else (and why wouldn't you?), it'll only be 6K.
That's the direction BOXX are trying to steer me and one I'm looking at. Their overclocked i7 and a dual Xeon RenderBoxx would be pretty rockin'. Sending off my company info for their "test drive" program now.
Nothing to lose!
Really? name them.
And if you consider the PPC line they did it with every single release.
Utter nonsense as it has nothing to do with Intel.
Can believe but as usual with you that has little to do with being true. There wouldn't be any TB peripherals if other folks weren't getting them. It was over 6 months before Apple even sold any TB peripherals (TB display in July 2011).
The stark reality was that there was practically nothing to hook to the Thunderbolt port for 6-9 months. There was going to be an intense system vendor race to be first for "the port to nowhere" ? Yeah right.
That doesn't mean other folks couldn't buy them. There are more than a few they didn't want to get burnt like Sony did:
http://www.theverge.com/2011/10/14/2490694/how-sony-accidentally-did-the-right-thing-with-light-peak
When Lightpeak shifted to being Thunderbolt, Sony had sunk R&D into a solution that died. That's isn't an "exclusive Apple contract/access". That is nobody buying it because the 'standard' is flaky.
Look at the track record to date. Even two years later the adoption by system vendors is still relatively low.
More strawman type nonsense? This is a very common a pattern for you. Anyway, I won't play. You have a web browser right?
That isn't always a good solution. Software licensing can be an issue. Even if the software allows for render nodes, it may only allow for the same thing to be rendered split across different boxes, rather than different frame ranges rendered on different nodes, and it may soak up time as the various nodes need to retrieve data. It also becomes more of a support headache trying to troubleshoot them if one happens to time out unexpectedly. Some software is terrible when it comes to notifications. It should also be considered whether we're talking about something like rendering in after effects or rendering in something like a 3d app where you might be able to split different passes between machines. Anyway, I just wanted to add that many details must be considered.
It doesn't have to always be a good solution. Just large enough to reduce what is left over to a relatively small (even for Mac Pro volume level) solution space.
if dealing with 3-5 workstations all using the same software already have a licensing "problem" with multiple deployments. A poor man's "batch job" is VNC ( Apple remote desktop facility) Login into the batch machine and start the just like the machine sitting in front of. The job doesn't have to be split fine grain. It isn't like a $6-8K Mac Pro is a wimp of a system. The files could be on this shared server in the first place so are being retrieved from disk... (not any different that being on a workstation with local disks. )
Yes, can still find corner cases, but the breadth of spaces where this works is pretty large. Alot of folks defacto do this now where the fire up a job on "Monster Mac Pro" and then while it is busy go do some other work on MBP or some other box.
That is only with the sophisticated "auto batch job scheduler" with network admin overhead.
Multiple projects have coarse grain parallelism aspects to them. For instance there may be ingest , compose/edit/debug , and export phases. Doing all three concurrently typically increases overall throughput. That's one reason why scaling up from single persons to multiple folks working on several things typically is effective if don't go overboard and segment the work skillfully.
Trying to peephole optimize some narrow subphases can miss the forest by focusing on a tree or two. And many vendors love those "ultra mega" configurations because they are typically priced non-linearly. That non linear gap can solve some of the additional software licensing issues. There are many details but the more critical task is to filter them into impactful or not categories relatively early in the process.
Alot of the discussions I see and the sales pitch gimmicks I see involve do attention focusing on peephole issues so that don't see holistic picture.
If they're not, they're not Pro Machines. The ability to modify a machine to meet your needs is key to people that make a living with their computers.
The first thing I do with every Mac I buy is replace the drive and memory (Makes Lion and ML even more annoying than they normally would be).
The day Apple dictates to me what I can and can't do, is the day they lose me as a customer. I don't buy things because they're thin or shiny.
I plan on buying a pro soon after the release ... not 'day of' in case there are any major problems that they need time to fix first. I am not need it for pro work though, more of 'advanced hobbyist', so waiting a few months does not hit me in the wallet.
My main concern is that it not lose upgradeability and become something a bit more expandable than my microwave. That it will be faster and with some new bells and whistles (USB3 and whatever) is a given. Just as long as they do not backslide on the whole concept of a tower.
My main concern is that it not lose upgradeability and become something a bit more expandable than my microwave. That it will be faster and with some new bells and whistles (USB3 and whatever) is a given. Just as long as they do not backslide on the whole concept of a tower.