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.
Mac Pro alternatives
- Thread starter starkrayz
- Start date
- Sort by reaction score
I think the biggest problem people have is that ATI 5870 goes for like $200 nowadays, yet Apple still charges $500 for it. For almost the same $, you could buy AMD 7970 which is a lot faster. It's not a bad card, just not worth the money Apple asks for it.
That's also true. It's not a $500 card any more.
If ATI started to include EFI ROMs on their cards, it wouldn't be as big an issue. But I doubt the OEMs want to carry that burden.
Are you referencing a "true" 6970 or the 6970m? The 5870 and the 6970m are in no way a "wash". It has lower clocks and almost half the stream processors. The 3dmark numbers are true showing the 5870 as twice as fast. In real scenarios it is obviously not 2x as fast in all cases but still beats it handily.
In demanding titles it can beat it by 50-100%. Even more so because on a Pro you can attach a resolution that you can actually play in native res with everything cranked. There are zero new titles that a 6970m can play native with everything on thanks to Apples huge 27".
Underpowered for the screen size, again, from Apple.
The specs:
6970m:
Stream Procs: 960
Texture units: 48
Core clock: 680MHz
Memory Bandwidth: 115.2GB/s
5870:
Stream Procs: 1600
Texture units: 80
Core clock: 850MHz
Memory Bandwidth: 153.6GB/s
REAL 6970:
Stream Procs: 1536
Texture units: 96
Core clock: 880MHz
Memory Bandwidth: 176GB/s
How are they in any way a "wash"?
Last edited:
I'm pretty sure he was talking about desktop AMD 6970. As you can see in the benchmarks, there are cases where 5870 is on-par with it.
Irritation...slowly...sinking back down
----------
You mean when IBM or Motorola made the CPU's?
True. Intel is working on improving along multiple dimensions. There are tweaks to the CPU throughput but the graphics were farther behind the competition. So it made sense to give a heavier weighting of the increased transistor budget to graphics since there are far more folks bad-mouthing that aspect than the x86 cores.
The TDP is a general trend for everyone except for the upper end GPU (which seem to be getting hotter over time ..... which long term is going to push them into smaller niches. )
But Sandy Bridge is a architecture shift, "tock", that is expected. Ivy Bridge is primarily a shrink, "tick" , with some minor arch tweats ( bigger cache , refinements that didn't make it earlier, etc.). One reason Intel graphics was relatively slow was because it was small and on strict power budget. Intel doesn't have to do a arch major shift in Ivy Bridge graphics just add "more resources" because each computational unit is smaller and uses less power.
Irritation...slowly...sinking back down
Yeah, desktop 6970. Comparing the 5870 with the mobile line isn't worth my time.
Was it similar under Windows? Gpu driver optimization always seems to suck on Macs. I also wish we had the 10 bit out option AMD has on their FirePro cards on Windows. The firepro version of the 5870 has it, and 10 bit display port has been out for some time. It's not that everyone would benefit immensely from it, but rather it would be nice if Apple at least kept pace with some of these Windows side features on OSX.
Other oems do this to a lesser degree. Most of them have scaling pricing to some degree as in updated frequently based on the cost to build a system or perform such upgrades. The Quadro 4000 costs more from Apple too. There's a PNY Quadro 4000 Mac edition. I'm not sure if it's a different card, but it seems to have fewer complaints than the one from the Apple page in spite of being labeled "Mac edition" (it's around $750 with a longer warranty rather than $1200). My complaint is the same on the 6 core mac. Cpu cost drops by roughly half, computer price remains consistent. It makes the idea of buying one so late in a cycle really annoying.
Are you sure he didn't just mean pre i-gadget Apple?
Definitely. Ivy Bridge is actually a "tock" for the graphics because the architecture is brand new. On top of that, the EU count is increased to a maximum of 16 and some mobile chips carry higher Turbo than SNB.
To be honest, I don't know where the IVB CPU performance increases come from. The cache speeds and capacities are the same (yes, even L1 and L2). The Turbo speeds are similar as well, although it's possible that IVB will be able to retain the Turbo speeds for longer periods.
The benchmarks I looked at were all under Windows.
OS X supports 10 bit output. Not sure where the breakdown is on that. I've seen it working on a 4870 under OS X, but I heard the 5870 isn't working. Seems like an ATI driver issue.
Apple is pretty restrictive on drivers, and overall yeah NVidia drivers tend to be superior to ATI, but ATI has been workable on Windows. I have yet to see any signs of driver support under any card. I never heard of it happening with the 4870.
http://forums.adobe.com/message/3286011
If you read a bit down that thread, it mentions a couple issues. I still get the feeling that Apple views this as a minor portion of its users, so they don't really care. The silly thing about neglect is that they're neglecting a very stable portion of their users.
On a side note, I'm wondering when Sandy Bridge E will finally ship in volume.
What are the current restrictions like? I've noticed in a lot of comparative testing, the tests don't really run for any extreme length of time. It makes me wonder how long these machines can keep cranking at maxed out turbo settings (say you were rendering or processing a ton of photos or something, and yeah I'd prefer a mac pro for that).
Last edited:
I knew someone with a 10 bit monitor who ran with a Mac Pro and a 4870. So I know it's possible.
Other reports of 10 bit working with the 4870:
http://www.luminous-landscape.com/reviews/accessories/10bit.shtml
OS X doesn't support system wide 10 bit (i.e. you won't see it in the monitors prefs) but an OpenGL app can kick it into 10 bit mode. Photoshop's OpenGL mode supports 10 bit output.
Honestly, reading over that thread, the Adobe support guy doesn't sound like he has much of a clue. Not that Apple hasn't had 10 bit support issues, but he's dead wrong about a few things he said (or misleading.)
Bleck it still annoys me that these companies can't cooperate to make a workable solution here, and yeah I own one of those displays (in the link). It's sitting right next to an older NEC. Both are quite nice. NEC has more weird bugs. I kind of wish the Adobe 1998 displays hadn't taken over so quickly, as they do have some issues. It's really not difficult to work within sRGB. Then again I also hate working on displays with obnoxious contrast levels because it has a tendency to hide artifacts and other problems in both stills and video.
I'm in this exact same boat with my 1,1 Mac Pro. I'm sure there are many of us. I know that upgrading now (with a hex 3.33) will run rings around what I have, but I'll go nuts if I spend 5K now (machine + RAM), and a new Pro was released in a few months for the same price.
You are correct. Here is the graph I am referring to. It shows the decline in the desktop sales vs laptop sales since 2000.
It does not take a genius to figure out that they are selling way more mac mini's and iMacs than they are selling Mac Pro's. And even if the percentages were evenly divided among desktop models - does anyone think that the Mac Pro is the biggest selling Desktop Apple makes? Its hard to even imagine an upward graph the past several years in Mac Pro sales.
Anyway, I don't want to come across like some kind of hater, I am the opposite, I am a huge Apple fan - and beyond that, I want to see us have a big, powerful, upgradable desktop well into the future. But seriously folks, I don't see anything encouraging about the Mac Pro's future. Unless they sort of reinvent it, which I hope is what will happen - but I wonder if Apple leadership are looking at this market as a place they even want to be anymore, the graph if nothing else shows where Apple's growth is coming from and its not from the Mac Pro.
Here is the graph:
https://www.macrumors.com/2011/10/31/apple-questioning-the-future-of-its-mac-pro-line/
Last edited:
It would be interesting to see a similar type of graph for the likes of, say, Dell, or HP. I'll bet it's similar. Laptops have come a long way in overall reliability, usability, and desirability since that graph start (2000!)
Sure. Just like I'm sure Dell and/or HP are cranking out way more lower-cost desktops than they are their monster workstations. Very few of their customers end up purchasing the monster workstations, but some people do. So they keep making them.
The truth of the matter is: we can speculate till the cows come home. Until the dual-processor Sandy Bridge E Xeons are released in March/April, we're not going to know for sure what Apple's plans are. Hell, we might not even know at that point. My personal speculation is: we'll see new ones. Some people will run right out and buy them (woo hoo!) Others will bitch that they're too expensive. And in a year's time, there will be more gnashing of teeth that the Mac Pro line is in danger of dying due to a lack of updates.
It's a pattern that the Mac Pro line seems to follow time and time again. Study history: it's a good way to predict the future.
jas
The problem with that graph is it masks the fact that Apple's desktop sales have risen year on year - with the 2009 financial year being the exception. Their laptop sales have performed so well you have to look deeper.
The average price per desktop sale is down a lot, which likely indicates Mac Pros make up less of the number, but $6.5b over 4.7m desktops leaves a lot of room for Mac Pro sales to have easily been anywhere from 10,000 to 250,000 units in the 2011 financial year for all we know.
I personally believe sales are down quite a lot. The majority of potential Mac Pro buyers aren't going to be idiots or sheep. They are going to be considering value for efficiency and the Mac Pro has dropped off a lot since the 2008 models, even more so as other systems have moved to quad-core CPUs and support more memory. There is also the life span of Mac Pros to consider. It isn't all doom and gloom though, in the favour of increased sales we have more people than ever creating content professionally and Apple's numbers continuing to increase. People still value efficiency - as you can see from PC workstation sales continuing to rise as desktop sales decline.
End of the day, for all the speculation we have on here Apple have done nothing out of character since they switched to Intel in that they wait for Intel's workstation platform. The 2010 delay was a little longer and perhaps due to getting rid of older stock first or because they didn't want 2 years between models, but it wasn't a massive change.
The Mac Pro will likely end at some point, but usually what people see as evidence is nothing out of the norm - they just don't look at how Apple have behaved in the past.
How about lowering the price tag? There are claims here that the similar systems can be built for half the price.
No it doesn't . It shows the percentages of desktop vs. laptop sales. For each year how what was the allocation of those sold. It shows absolutely nothing about unit sales. The units from each year are normalized to be what indicates 100% for that year. There is nothing here that illustrates how many are being sold between years.
One of the underlying factors in the graph is that the growth rate of laptops has been much higher than desktops. The total number of Macs sold has been going up every year. Characterizing that as a "decline" is misguiding.
If you actually look at a graph of units sales (instead of normalized percentages). For example
http://gigaom.com/apple/the-ongoing-decline-of-the-desktop-mac/
[the FY2011 numbers are up also for desktops if IIRC. ]
At this point it should be obvious that desktops have been generally increasing for the last couple of years.
Correct. That is the core Mac Pro problem. If laptops are increasing at 16-25% percent every year and mini's/ iMacs are increasing at 10-15% every year, then even if the Mac Pro showed 0-1% growth it would be in trouble. Negative growth would put it in the "candidate for cancellation" list.
I think Apple isn't so dense to not to know that there are hiccups in the graph due to this longer than usual upgrade cycle on parts. I suspect that is why they will do a "wait and see" with one more release. If can show an sustained growth rate through most of the year they'll probably continue. If there is a burst from the folks waiting and then it drops back to very low (or negative) growth then they'll axe it. So the core issue is not what Apple is going to do. It is whether folks will buy them or not. If they don't, it will get canceled.
Last edited:
But that is at best 5% and at worse approximately 0% of desktop sales. As percentage of overall Mac sales at best down very near the 1% range.
The Mac Pro doesn't have to outsell the other models but it does have to "tread water" in terms of growth. When the execs see the weekly pie/bar chart of Macs sold and the Mac Pro disappears from view.... it will probably get axed. That's where Jobs' "Nobody is buying them" comments comes from. It doesn't have to be literally zero for that to be effectively true.
There never was a massive change in the previous upgrades where Intel didn't change the socket. Apple just tweaks what plugs into the board ( generally just like everyone else in the high end of workstation market).
They have done similar (in terms of features present on the motherboard) for other Mac models too when there is no socket change.
Last edited:
Yep, not saying that Apple sales have not grown over recent years. We are getting bigger and bigger market share and I love it.
The better Apple does, the more/better new products we can hope to see.I don't want to split hairs about what the graph does or does not say, etc. The point I am making is that the Mac Pro and desktops in general used to be 75% of Apple's sales in 2000 and now its down to what? 25% or less?
Whether that graph is the same for the rest of the world or not, it still indicates that the Mac Pro is becoming less and less relevant, even within Apple. And as a percentage of revenue as it shrinks more and more its going to be harder and harder to invest in it. Do they focus resources on the new Apple TV set which could be a whole other new influx of growth in a new market, do they focus on more innovation with the iPad to stay ahead of the competition, focus on MacBooks which are big sellers, or do they divert resources to a shrinking marketplace and increasingly irrelevant product like the Mac Pro? If what the article says about the Mac Pro not even being profitable anymore, I don't see it surviving. This makes me sad. I wish they would do something aside from just wait for the next Intel chipset.
You don't even see any marketing dollars or advertising spent the Mac Pro. They need a whole new strategy with it. We can sit here and try to defend the current state of things, that might make us feel better - but I still say things do not look good for the Mac Pro.
Interestingly enough, the graph you supplied is not about Mac Pro sales. While its no secret laptop and mobile devices are greatly distancing desktop sales its more then likely showing consumer numbers and not business and enterprise use.
While professionals, business & enterprise can make due with ipads & MacBook Pros & Air you will see an increase/increased need for desktops like imac, mac mini & Mac Pro's in this market. The use of Macs in this market is expected to increase by 57%.
Like some one mentioned, while Laptops have overtaken desktops, desktop sales have still increased not decreased, so they are still profitable and they are still needed.