Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Windows has nothing to do with it.

for clarity, i'm generally saying windows in reference to computers that aren't macs.. unless i specifically point out that i'm talking about the OS or software such as in your quote of mine

But yes, having more than double the amount of CPU horsepower does speed up the software.

i know this. i feel that i've made it clear in the above posts.
and i can see that you would like to have 50 cpus inside your work station as opposed to a streamlined hub and get your cpu elsewhere.

i just wish you could say it like it is instead of saying the computer is crippled and/or affects your work.. because as far as i can gather, it's cheaper, cleaner, quieter, and cooler(temperature and even the other kind of cool ;) ) to have a cpuasaurs in it's own dedicated location doing it's own dedicated tasks..

(actually, the way this conversation has ended up, i don't wish for you to 'say it like it is' or whatever.. as far as i can tell at this point, our disagreement lies in where the physical hardware should be located in order to have the desired power.. it's just differing logistical opinions which is fine with me.)
 
The essence of the Mac Pro is expandability.

This new machine is a really powerful Mac Mini. There may be a market for a really powerful Mac Mini.

I just hope that Apple also produces an new, powerful MP.
 
Many of those corners are illusory, because folks don't like the idea of detached storage.

As I said yesterday:

"Great! Oh, but wait there's no such thing as a free lunch: so for a small business who doesn't have their first fiber SAN, just how much will it cost to stand up a new turnkey system from scratch?

Insofar as capacity, let's say just 12TB worth, as that's an obvious comparison: the value performance baseline is $1500 (three 4TB internals in RAID0 on the 2012 Mac Pro).

Beat the price and you have a viable business case. Or fail...and you're just like any other IT Dept who has forgotten that they're a support organization who lives on overhead dollars paid for by the business units."

Frankly, I can live with internal or external ... but the decision still boils down to the brass tacks of the business case, which is always going to ask: HOW MUCH DOES IT COST?

I'm not going to write a blank check.


-hh
 
Last edited:
hey, i'm not rich either.. over the years, i've accumulated 2.1TB storage (not counting the ssd or portable drive).. so 3 internals and one external drive..

i spent about $900 on those drives.

if i spent the same amount today, i could get 8TB storage via thunderbolt..

Time is an illusion (lunch time doubly so): back in 1986, I spent $2200 for 20MB ... for a Hyperdrive in a Mac Plus. Today, I'm given vastly larger capacity USB thumb drives at Trade Shows...for free!

so, from where i'm sitting at least, this whole 'thunderbolt is crazystupidexpensive' thing is a fallacy.. i mean, i can -right this minute- go buy twice the storage for half the price of what i've already spent.
you see that, right?

The problem is that your comparison is skewed. Now let's look at a today-vs-today comparison with your $900 that bought 8TB on Thunderbolt.

Today's pricing on Newegg has the 4TB Segate (ST4000DM000) for $179.99 That means that $900 will buy five of them. That's 20TB, and it can all be installed internally on a 2012 Mac Pro.

Granted, we can try to ankle-bite pedantic minutia, such as if we have to spend another $19 for a mounting tray to go in the current Mac Pro's second optical bay, but the above might not include the $29 to pay for its TB cable either. Ditto for if both are Enterprise Class or consumer stuff.

The bottom line is simple: given the same price point (etc "all other factors equal"), do you consider 8TB or 20TB to be the better value?

YMMV, but I kind of suspect that the 20TB is a teeny weeny bit of an obvious "better".


-hh
 
i guess Apple engineers are amateurs because there are a lot of experts here shooting holes in what they have done.
 
The problem is that your comparison is skewed. Now let's look at a today-vs-today comparison with your $900 that bought 8TB on Thunderbolt.
[...]
YMMV, but I kind of suspect that the 20TB is a teeny weeny bit of an obvious "better".

hey hh
yes, of course i see what you're getting at and yes 20TB is the better value vs 8TB when comparing that way etc..

but yet again i'd like to point out that i'm talking about my personal experience.. i would say i 'need' 2TB right now (and that's what i have).. and if i were to buy 2TB along with a new mac, i would feel like i'm making out great.. i mean, what, spend $50-60/year for the next 4-5 years on storage? that's chump change, you know..

if you're concerned that i'm being ripped off for spending too much on a drive which will work with my choice of computer then thanks, but i'll think i'll be a-ok..

(but hey, if you ever see somebody robbing me on the street for $50, please feel free to chase after them ;) )


or- if you're simply trying to point out that on paper, it costs more to use macs when compared to other computers (i.e.- other computers are the better value), then thanks but i'm pretty sure i and most other people have already figured that out a long time ago.
 
i guess Apple engineers are amateurs because there are a lot of experts here shooting holes in what they have done.

Yes, sad that they weren't as clever as those Asus engineers who got TB working through a GPU of your choice AN ENTIRE YEAR AGO.

Well, hopefully they keep trying and don't give up despite their poor results thus far.
 
Yes, sad that they weren't as clever as those Asus engineers who got TB working through a GPU of your choice AN ENTIRE YEAR AGO.

I'm sorry but re-routing the screen output with a cable back to the motherboard is not clever, it's a hack at best.
 
i guess Apple engineers are amateurs because there are a lot of experts here shooting holes in what they have done.

Only if you're also assuming that Engineers always get the final say for any new product.


hey hh
yes, of course i see what you're getting at and yes 20TB is the better value vs 8TB when comparing that way etc..

And a lot of other ways too: one can also say "if I need 8TB, how much would it cost?" and the answer would be $360 (vs your $900).


but yet again i'd like to point out that i'm talking about my personal experience.. i would say i 'need' 2TB right now (and that's what i have).. and if i were to buy 2TB along with a new mac, i would feel like i'm making out great.. i mean, what, spend $50-60/year for the next 4-5 years on storage? that's chump change, you know..

Yes, we can ignore ankle-biters...although a pedantic FYI: $60/year * 5 years = $300, which isn't quite enough money to buy even one LaCie 2TB Thunderbolt drive.

The real question is ... at what point does the difference in costs no longer qualify as a "chump change" ankle-biter that isn't particularly important?

Yes, I know that this is beyond the very narrow scope of your personal use case.

if you're concerned that i'm being ripped off for spending too much on a drive which will work with my choice of computer then thanks, but i'll think i'll be a-ok..

Sorry, but I don't really care about your use case, because I have my own to care about. But getting back to the question: at some point, the lifecycle expense differences are going to be big enough such that it won't be able to be trivialized as 'chump change' by the consumer base.

For example, a use case capability requirement of 20TB would be 10x your "chump" number, which would be $600/year. Is that still a trivial amount?

Yes, this is a YMMV and it depends on many factors.

And granted, there is some value-added differentiation afforded to Apply by OS X, but it only goes so far and as the "Apple Tax" becomes more financially painful, ultimately, everyone has their breaking point.

Apple's design decision is pushing some of their existing consumer base further away...and statistically, some of them will defect. That's a business downside risk that should be offset by some upside gain potential. Question will be if the gambit works to Apple's benefit or not, and where is the growth opportunity ... or *is* there even one? This could be just a cost-cutting move with no intention or expectation of customer base growth.


-hh
 

hey hh
i had some stuff typed up regarding your last post but decided against posting (nothing 'bad' or anything).. my responses came out as if i were arguing you in a similar way i was arguing (what i feel is) misinformation by others in the thread.

i mean yeah, i understand what you're saying and agree with most if not all of it.. definitely- without a doubt- we all have our breaking points and with things like this, there are many factors which will contribute to and affect us individually on different levels.
 
Even with a maxed out iMac, Nodejs compiles still can take forever... so...

All this talk about Disk speed not being helpful... what pro isn't pegging disk speed? Or maybe "pro's" are mostly design pros or light video users. I worked at an Enterprise software company and though we built windows server tools, everyone used a mac. Compiles still took minutes, and doing 15-20 an hour is painful. Plus, you throw in running a few VM's at the same time, and a spinning drive would literally run non stop.

Are people not running 2-3 VM's, compiling C# and Node projects, and also having to do the design work at the same time?

It is a pro machine. I am surprised at how few enterprise software developers there are here.
 
Even with a maxed out iMac, Nodejs compiles still can take forever... so...

All this talk about Disk speed not being helpful... what pro isn't pegging disk speed?

hey sean.. i've been saying stuff about disk speed but not in the context you're speaking (i've been speaking in the context of batch data transfer as opposed to the working drive (and fwiw, i do have ssds in all my computers so i obviously see some importance there in my own work))

as far as i can tell, this new mac pro has the fastest disk ever in a mac pro.. along with the fastest graphics (by far), the fastest processors, the fastest ram, the fastest everything.. even wifi.. i mean to me, this thing screams compared to past mac offerings..

but so many people are complaining how slow and unusable it is (though no one has yet to give a real world reason) so maybe you can.

when you're compiling, are you generally working off the boot drive or is the data coming in from somewhere else? (because it seems as if everyone in this thread with thunderbolt complaints are talking as if they don't work off the main drive so it's thunderbolt causing the slow down).

anyway, with the type of work you're doing, do you foresee external drives via thunderbolt causing any slowdowns in the actual work?
thx


Are people not running 2-3 VM's, compiling C# and Node projects, and also having to do the design work at the same time?
if this is a poll then mark me down as not doing any of that except for design work :D
 
As I said yesterday:

"Great! Oh, but wait there's no such thing as a free lunch: so for a small business who doesn't have their first fiber SAN, just how much will it cost to stand up a new turnkey system from scratch?

Insofar as capacity, let's say just 12TB worth, as that's an obvious comparison: the value performance baseline is $1500 (three 4TB internals in RAID0 on the 2012 Mac Pro).

Beat the price and you have a viable business case. Or fail...and you're just like any other IT Dept who has forgotten that they're a support organization who lives on overhead dollars paid for by the business units."

Frankly, I can live with internal or external ... but the decision still boils down to the brass tacks of the business case, which is always going to ask: HOW MUCH DOES IT COST?

I'm not going to write a blank check.


-hh

Hm, you need 12TB of striped storage using traditional disk and accessible over a high speed connection? Under $1500? The LaCie 5big can do 10TB over Thunderbolt, which would be plenty of bandwidth for most SMB applications as direct attach storage. If you want that over FC, you're going to have to go rogue and build your own box, since any reputable vendor won't be under $2700 for a FC SAN solution, or $6000 for a turnkey setup with drives.

Here's your problem: a turnkey solution from a reputable vendor is going to cost some serious dosh up front, with the idea that because said centralized storage is used by multiple business entities, that it's worth the steep initial price. If you want a storage array for under $1500, I can easily build you a 40TB array with RAID 0/1/5/6/10/50 support. It won't have the warranty, support, or feature set of a proper SAN, and it won't be turnkey, but you'll get what you want at the price you're willing to pay.

That's neither here nor there, though, since this is ultimately about the Mac Pro. Will it force organizations to rethink their current setups? Absolutely. Will it put a strain on SMB's? You betcha. Thing is, though, Apple has always been a premium brand, even when they made Enterprise servers and SANs. You always paid a price premium compared to their competitors, and you likely always will, especially in their higher-end gear. I work for a Medium-sized business now, and we're facing a complete overhaul of our publishing division since Apple is pushing Thunderbolt so hard, and our old displays and FW gear becomes unreliable or unsupported. Do we like the prospect of replacing all of our old CCFL Cinema Displays with Thunderbolt ones at $1k a pop? No, but we understand that's the future, and that in order to remain competitive, it's an expense we'll have to endure, and the same goes for the nMP. We don't like it, but we understand that the long term picture of adopting Thunderbolt will be the increased flexibility available to us in terms of expansion, as well as the potential for reduced cost if Intel every succeeds with implementing it into PCs.

If the nMP is too expensive a proposition for you in both the long term and short term, then you either need to reevaluate your business expenditures if they're truly necessary items, or look to alternatives that fit your budget.
 
It's actually about the same price on MP5,1 as it is on MP6,1 when you do the sane thing of connecting your rotational RAID0 to the USB3 connectors as most normal people will do. So this whole discussion side-thread is moot IMO.

HOW MUCH WILL IT COST?

Option One:
2 - Three or four bay USB3 enclosures at $75ea: $150
6 - Seagate 3TB Barracudas at $125 ea: $900
2 - USB3 cables at $5 each: $10
--------------------------------------------------------------
Totals:
Price $1060
Total Capacity: 18TB
Top speed: between 900MB/s and 1GB/s​



Option Two:
3 - Two bay USB3 enclosures at $50ea: $150
6 - Seagate 3TB Barracudas at $125 ea: $900
3 - USB3 cables at $5 each: $10
--------------------------------------------------------------
Totals:
Price $1070
Total Capacity: 18TB
Top speed: between 1GB/s and 1.2GB/s​
 
Even with a maxed out iMac, Nodejs compiles still can take forever... so...

All this talk about Disk speed not being helpful... what pro isn't pegging disk speed? Or maybe "pro's" are mostly design pros or light video users. I worked at an Enterprise software company and though we built windows server tools, everyone used a mac. Compiles still took minutes, and doing 15-20 an hour is painful. Plus, you throw in running a few VM's at the same time, and a spinning drive would literally run non stop.

Are people not running 2-3 VM's, compiling C# and Node projects, and also having to do the design work at the same time?

It is a pro machine. I am surprised at how few enterprise software developers there are here.

Some people just want to post cute puppy pix on Facebook and uplaod an occasional video to Youtube. Perfect machine for it. At a reasonable price, they'll sell like hotcakes to the Semi-ProSumer crowd.
 
Hm, you need 12TB of striped storage using traditional disk and accessible over a high speed connection?

What I need is a price quote to substantiate the claim that remote SAN storage is a viable business decision for the small business centric use case.

Under $1500? The LaCie 5big can do 10TB over Thunderbolt,

Unfortunately, that's not a SAN over Fibre Channel, which was what the claim was touting as the glorious future for everyone.


If you want that over FC, you're going to have to go rogue and build your own box, since any reputable vendor won't be under $2700 for a FC SAN solution, or $6000 for a turnkey setup with drives.

So this "better" solution is 2x - 4x more expensive than the Status Quo. That's all that I need to hear.

Here's your problem: a turnkey solution from a reputable vendor is going to cost some serious dosh up front, with the idea that because said centralized storage is used by multiple business entities, that it's worth the steep initial price.

Please don't condescendingly assume that I don't comprehend the business use case of utilizing a centralized resource to effectively cost-share across multiple seats.

What you've now reluctantly verified (with prices) is that centralization doesn't always turn out to be cheaper.


It's actually about the same price on MP5,1 as it is on MP6,1 when you do the sane thing of connecting your rotational RAID0 to the USB3 connectors as most normal people will do. So this whole discussion side-thread is moot IMO.

I agree, which is precisely why I asked for a price for the "SAN!" claim: it has now been confirmed that it isn't anywhere close to being a cost-effective option for many individuals & small businesses.

FWIW, I won't contest if this represents "most" Mac Pro customers or not, as we simply don't have the hard data to substantiate it either way. But I do personally suspect that this is true.


-hh
 
It's actually about the same price on MP5,1 as it is on MP6,1 when you do the sane thing of connecting your rotational RAID0 to the USB3 connectors as most normal people will do. So this whole discussion side-thread is moot IMO.

HOW MUCH WILL IT COST?

Option One:
2 - Three or four bay USB3 enclosures at $75ea: $150
6 - Seagate 3TB Barracudas at $125 ea: $900
2 - USB3 cables at $5 each: $10
--------------------------------------------------------------
Totals:
Price $1060
Total Capacity: 18TB
Top speed: between 900MB/s and 1GB/s​



Option Two:
3 - Two bay USB3 enclosures at $50ea: $150
6 - Seagate 3TB Barracudas at $125 ea: $900
3 - USB3 cables at $5 each: $10
--------------------------------------------------------------
Totals:
Price $1070
Total Capacity: 18TB
Top speed: between 1GB/s and 1.2GB/s​

I'll take a half order of Option one with four enclosures for starters please. :D
 
What I need is a price quote to substantiate the claim that remote SAN storage is a viable business decision for the small business centric use case.

Well, if all that's needed is gigabit ethernet shares then the two options that come to mind are USB3 to EtherNet media-share hubs where you can essentially plug in any USB3 device and share it through that - connected to your station network hub. These start at around $50 and go up to around $250 or so. The other way is simply to use the MP the devices are connected to and share the volumes - binding both nics from each of the machines which have dual nics will help here a bit (200MB/s instead of 100MB/s). The cost for this later option is only the price of the extra LAN cable needed per machine.

If you need faster than 200MB/s which the bound nics can deliver then the price skyrockets astronomically with all the options I know of. But I dunno if TB can be used for networking or not. If it can then that's the solution to go for there.

I agree, which is precisely why I asked for a price for the "SAN!" claim: it has now been confirmed that it isn't anywhere close to being a cost-effective option for many individuals & small businesses.

FWIW, I won't contest if this represents "most" Mac Pro customers or not, as we simply don't have the hard data to substantiate it either way. But I do personally suspect that this is true.

Storage sharing is basically free after the cost of the storage devices themselves. Now, how much bandwidth you need to serve it up by will matter a lot in that regard. Traditional solutions of 4Gb/s and higher are very costly and probably not a good candidates for "individuals & small businesses".

I'll take a half order of Option one with four enclosures for starters please. :D

Hehe, yeah, the MP6,1 has only 4 dedicated USB3 ports. It would be nice if it had 6 or 8. :)
 
First, Thunderbolt doesn't permit two PCs to be networked together, so hooking up your laptop to a Mac Pro via TB simply isn't an option: another (different) cable will be required.

(going back to this as it's pretty much my only concern regarding the new mac)

is this limitation inherently impossible with thunderbolt technology or is it more likely that the capability isn't hooked up yet?

-ie.. is it possible (or likely/improbable) that we will see this ability in mavericks?
 
First, Thunderbolt doesn't permit two PCs to be networked together, so hooking up your laptop to a Mac Pro via TB simply isn't an option: another (different) cable will be required.

Wait a tick, is that really true ?
You can't run a wired network between Macs via TB ?
Precious .
 
Wait a tick, is that really true ?
You can't run a wired network between Macs via TB ?
Precious .

Where has it ever been said that can run network connection over thunderbolt cable directly.

MBA and rMBP don't have Ethernet built in for which apple offers thunderbolt to Ethernet adaptor if don't use the USB to Ethernet instead.

Are you thinking of FireWire where could do network.
 
Well, if all that's needed is gigabit ethernet shares ...

And anything that can run ethernet satisfies this...and Gigabit was standard on the G5 a full decade ao.

If you need faster than 200MB/s which the bound nics can deliver then the price skyrockets astronomically with all the options I know of.

Yes, this has been the underlying point on the "loss of internal bays" discussion splinters: this level of performance capability that's been baked into the Mac Pros for years, and configrations can take it up by another 50% without major grief. The "least bad" option today appears to be to use USB3...something that really should have been OEM on the 2012 via a PCIe card, but that's a tangential gripe of Apple's chronic neglect of the MP platform.

But I dunno if TB can be used for networking or not. If it can then that's the solution to go for there.

Depends on what you consider to be 'networking': TB can be used in target disk mode, and TB can be used with an Ethernet or Fibre Channel adapter too.

....but the idea of hooking up two autonomously booted/managed Macs with just a TB cable and having it work like Ethernet? Not so per Apple marketing, nor even any well-informed discussions on the underlying technology.

FWIW, I would like to see it because it is an interesting idea from the perspective of any easy-DIY cluster for modular performance growth...of course, we could also see Apple not wanting that because it slows/reduces hardware upgrade cycles.


-hh
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.