Separate names with a comma.
Discussion in 'Mac Pro' started by JesterJJZ, Feb 23, 2014.
Haven't seen it posted here yet.
Looks like a Geiger counter.. (and that's a bad thing)
Its as ugly as their cameras. I wonder how little research went into if the case will impede heat dissipation, and also if its fan is as loud and awful as those found in red cameras themselves. Cool idea though.
Looks better and more functional if you as me. Kinda looks like a Ghost Trap
and that's awesome!
Can someone tell me why they need 104 screws to hold it together?
Yeah this has been posted before.
I'm not sure I like it though.
perhaps we should just wait for someone to make a custom case that includes the possibility of adding Thunderbolt disks, including Thunderbolt to PCIe and other stuff.. and then it kinda looks like:
Looks like a giant pencil sharpener.
I like its industrial style though.
I wonder if the outer cylindrical case of the nMP has been removed too and what impact that would have on airflow.
this is an interesting concept. taking the guts of the nMP and making a new case for it. if one had the time and capital, you could make a slightly taller case, keeping the same design cues (polished cylinder), and add a couple HD bays in the bottom (providing they fit) and maybe even an optical drive. could eliminate clutter and keep things as neat looking as we all like them to be (for those that want/need those things).
Where do you get that from. The quote under the picture is.
" .. My own little machined aluminum box with a TrashCan Mac Pro stuffed inside plus ... "
The 'stuffed inside' is highly the Mac Pro is inside this larger container. Other commentary on facebook (if chase back the original photo) suggest that it is just the Mac Pro turned horizontal and the ports coming oriented to the side. [ since the creator didn't like the "ports on the back"].
The blower on top is suggestive of trying to keep both the Mac Pro and the additional equipment ( ".. plus ntegrated REDMAG MINI reader with 8TB internal Thunderbolt RAID ..." ) cool.
I suspect this would be less interesting to many if the cabling required was also shown instead of completely clipped out of view.
Can be taller but if don't move/adjust the center of gravity it would also be more prone to tip over also. Simply cranking on a single dimension most often brings tradeoffs ( decrements as well as benefits ).
In order to contain the original Mac Pro starting from, it would need to be wider and taller.
If rip off the cover going to screw with the flow of air over the "outside" of the internal boards as well as the inside. Crank up the Power supply and will need a different volume.
Sure can keep throwing money at the problems, but the market is going demand along the lines of
which is a bigger container that simply just put the smaller Mac inside of to hit any sort of broad base value proposition that has traction.
How things work is an important aspect of design as much as how they look. Turning a heat chimmney horizontal is an impact on how it works. Same with just willy nilly wrapping a container around a Mac just to hide cables.
I'd buy it. Would match the Epic's and get rid of some cable mess on set.
please, sir, pardon me for having an imagination. my verbal ability to express my nonsensical ideas are not as good as perhaps my visual ones. willy nilly? yessss. viable? maybe.
please, allow me to try to better describe what i had in mind in a more visual, albeit crude, manner:
so, as you can SORTA see, i wasn't talking about stuffing the whole mac pro into a new case, but machine to precision a new "case" for it. obviously it needs a lot more thought than I put into it, certainly, but with some (tons of) "time and capital" i think it could be a possibility. it would be taller but i hardly think the weight of three drives and a power supply could topple this to merit a wider design. plus i'm betting not a lot of people place $3000+ dollar equipment on anything less than a sturdy surface. the miniserver isn't viable (to some) due to shear processing power, however, you may be right on the VALUE of such a thing.
BACK: two HD (or SSD) drive sleds and one optical drive sled with handles. thunderbolt port on back to connect to mac pro (a short cable unseen from the front) wasn't sure if it'd need 1 or 2 ports to run 3 devices (guys, i'm not THAT smart). otherwise, access ports and lock mech would line up the same as stock shell. would probably need a power port somewhere as well. probably.
FRONT: same as stock finish (or as close as a guy can get) but with a slot-loading slit at top.
XRAY(ish): you would just slide this over the top of the mac pro, connecting it with 1 or 2 TB (and power) cables. dimensions/proportions are probably off, but you've got the 3 drives and also a helper fan at the top of the new shell to keep that air, with the help of hot air physics, moving up, up and away. you would also need to engineer a power supply in there, or perhaps an outside brick in the power line.
again, this is by no means perfect and quite possibly impossible, but you're suggestion that it is not a viable or yet a desirable design is an opinion i do not share. a lot of work and money to hide cables? 100% YES. agreed. Aesthetically pleasing? That's to be debated. I think I could live with something like this though.
The problem with your design the the nMP will fry your HDD and disk drives. Those stuffs need to be placed at the bottom of the nMP and also to have vents to allow the nMP to suck in air from the bottom
i guess that's why i'm not an engineer. just a guy with photoshop. however, if i was given all the money apple has, i'd do my best to prove you wrong. now THAT is my stubbornness at it's finest.
In appearance and intended functionality it is a cool idea.
As a potential product design it has a major flaw. The problem is heat dissipation.
This design puts an extra 450 watt heater in the base of a storage enclosure. The hot air exhausted by the mac pro will be used to cool the drives.
I do like the look of it, and I like the thought behind what you are are trying to do. However, from an engineering perspective, I think it is not worth trying to do. It is trying to fight physics rather than working with it.
BTW: I have similar problems with the idea of integrated storage beneath the mac pro. Why add another 100-200watts of waste heat for the mac pro to deal with?
I think thermal principles are over-exaggerated. As long as their is airflow, it will be fine. After all, the whole case is a heatsink....
I have 15x3TB drives in a RAID6 array in a cupboard just big enough to house them with two small 80mm fans on the rear wall circulating air. Fans undervolted to be quiet.
Our ambient air temperature can exceed 45'C. I have never had any problems with this setup despite it running 24/7
I think what you propose in your mockup is trivial by comparison and seems like a valid concept to me
But then I'm a doctor and my degrees lie in fields other than thermodynamics...
It's too bad Harold Ramis didn't live to see it.
if i wanted to fight physics i would be pointing the helper fan downwards towards the mac pro which would be ridiculous. i'm [attempting] to work WITH physics since (a) hot air rises and (b) the helper fan would only help the vacuum created by the tube to pull hot air out.
also - agreed - putting these things below a mac pro would be equally ridiculous.
i'm not buying the HYPOTHESIS that the heat from a mac pro would ruin two drives above it. hard drives have been proven to literally take a beating in high heat situations. and it's not like the air would be stagnant, it would still vent out the top. consider that GOOGLE has provided a very detailed report on failure trends of magnetic media from 2007 where they have concluded
"it is likely that there are other effects which affect failure rates much more strongly than temperatures do" - i'm betting they own a lot of drives and have a lot of experience in that field.
just wait and see, when I marry a supermodel, I'll take all of her money and vest it into R&D for this. just you wait....
You are operating within design tolerance for most drive manufactures who typically place an operating ceiling of 60C (140F) on their drives. 45C is 113F.
Drives don't suddenly fail at some magic temperature, it is not like a melting point or other state change with a fixed boundary. It is a change in a probability distribution.
Drives, as they age, simply have higher rates of failure at elevated temperature. At around 3 years, the probability of temperature related failure is expected to be 3 times higher at 45C than at 35C.
You are using raid 6, capable of tolerating a dual drive failure, which is great. This will help mitigate your risk. I would suggest, that when rebuilding after a single drive failure that you leave the cabinet door open. Also, I would suggest replacing drives after recoverable SMART failures since they are much higher predictors of impending failure than is temperature.
Also, you state: "As long as there is airflow, you will be fine". Air flow is only part of the model. Heat transfer depends not only on flow, but on delta T, the difference in temperature between the cooling medium and what is being cooled. That is why raising the temperature to the air intake (which is the Mac Pro exhaust) makes such a difference to this design.
That paper reveals more than you imply. In their study, temperature alone, was a much poorer predictor of drive failure rates than shown in prior studies. An earlier study by a Seagate engineer in 2000 reported "that MTBF could degrade by as much as 50% when going from operating temperatures of 30C to 40C". The Google results were surprising because: "In our study, we did not find much correlation between failure rate and either elevated temperature or utilization."
Part of this discrepancy can be explained by advances in both drive design and manufacturing over the intervening 7 years. Another part of the difference appears to be due to the fact that google identified several sources of data which show much higher correlation to failure rates than temperature such as, SMART data, sector and seek errors, etc. Finally, the role that temperature plays is not uniformly distributed over drive lifespan.
Even though temperature alone was a poor predictor of failure, it was still statistically significant. At ages less than 2 years, temperature alone was insignificant, but at 3 years the failure rate attributable to temperature double when temperatures rise from 35C-40C, and triple when comparing 35C-45C. So, operating temperature still matters for older disk drives.
At this particular point in time, I think this is highly relevant because people migrating from older Mac Pro would use such an enclosure to migrate internal storage. If so, the drives are likely to already be 2 or more years old.
The optical drive is a greater concern than the spinning hard disks. These are much more sensitive to proper cooling. Most venders rate their drives for use only up to 35C (95F). Cooling and heating cycles for DVD and CD disks also impair the optical media.
Working against physics is far more than "heat rises". It has to do with the fact that heat travels from hotter to cooler temperature regions, and does so at a rate which depends on the difference in temperature not the absolute temperature.
When these external drive are under idle conditions or modest load, the exhaust from the Mac Pro is likely to be adding heat to your storage devices instead of letting them dissipate heat to the much lower ambient temperature of the room. This is working against physics.
When the Mac Pro is at high load, the exhaust may be at higher temperature than your optical drive is designed to function. No mater how much air flow you get, it will be heating the drive beyond its operating range rather than cooling it at all. This, too, is working against physics.
Beyond this, even though your design calls for an additional cooling fan, it runs in series with the Mac Pro's existing cooling system rather than in parallel. This means that when the Mac Pro is relatively cool, but the components above need greater air flow, the system below will be acting as drag: impeding air flow rather than accelerating it. This requires an even larger and beefier fan operating at a higher duty cycle.
In over 20 years as a systems engineer my most useful rule of thumb was, "If a problem seems really difficult, STOP and look again: you are probably trying to solve the wrong problem."
In this particular case, I don't think that your proposed design is one that is worth throwing a bunch of money or time at. The design pre-supposes a high thermal load which is independent of the problem being solved.
Adding a Mac pro inside a storage enclosure adds a large cooling problem to the design. From a physical standpoint your proposal says "To start with, let's quintuple the thermal load." As an engineer, this is tantamount to saying let's take a small problem and make it large. Simplicity in industrial design emerges from solving a few problems well. This is unlikely to emerge when the initial constraints add new problems which did not originally require a solution.
You obviously have the graphical design chops and creativity to produce some nice ideas, and you should be encouraged to continue. However, there is a huge gulf between working with pixels and working with objects. The real world has real constraints. If you want to transition from graphical design to industrial design, you should strive to gain some more experience with, and understanding of, physical limits.
I realize that this critique might be interpreted as pissing on your cheerios. That is not my intent. I hope that you can see it in the light in which it was intended: an attempt to get you to look at some issues which you may be unaware, and which pose larger problem than I believe you thought.
I wish you well.
thanks for that. i'll stick to pixels.