D510MO (Mount Olive) : 13watt Atom D510 mITX board

All about them.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
aztec
Posts: 443
Joined: Mon Dec 12, 2005 5:01 am
Location: Foster City, CA

D510MO (Mount Olive) : 13watt Atom D510 mITX board

Post by aztec » Wed Dec 09, 2009 3:45 am

http://www.tranquilpc-shop.co.uk/acatal ... ount_.html

For ~$99 and passive at that!

This could be a great platform for a simple WHS box.

andrewb
Posts: 5
Joined: Mon Sep 15, 2008 4:49 pm
Location: ca

Re: D510MO (Mount Olive) : 13watt Atom D510 mITX board

Post by andrewb » Wed Dec 09, 2009 11:47 pm

In the US, avaialble here, http://www.mini-box.com/D510MO-mini-ITX-Intel, also cheaper.

Combine that with a picoPSU-80 and a M350 fanless enclosure and you got yourself a fanless PC with the latest Atom processor....

Finally, Intel is getting it right...

-A
aztec wrote:http://www.tranquilc-shop.co.uk/acatalo ... ount_.html

For ~$99 and passive at that!

This could be a great platform for a simple WHS box.

aztec
Posts: 443
Joined: Mon Dec 12, 2005 5:01 am
Location: Foster City, CA

Re: D510MO (Mount Olive) : 13watt Atom D510 mITX board

Post by aztec » Wed Dec 09, 2009 11:55 pm

andrewb wrote:In the US, avaialble here, http://www.mini-box.com/D510MO-mini-ITX-Intel, also cheaper.

Combine that with a picoPSU-80 and a M350 fanless enclosure and you got yourself a fanless PC with the latest Atom processor....

Finally, Intel is getting it right...

-A
aztec wrote:http://www.tranquilc-shop.co.uk/acatalo ... ount_.html

For ~$99 and passive at that!

This could be a great platform for a simple WHS box.
Exactly!

Will be selling my k45 mobo, Celeron 430 and PSU.

Already picked up a Pico, just waiting for this board now.

Nice little WHS box. :D

sousa
Posts: 20
Joined: Fri Jan 27, 2006 2:44 pm

Post by sousa » Thu Dec 10, 2009 7:28 am

Here's a pseudo-review where they measured power consumption,
http://cartft.de/support_db/support_fil ... iew_EN.pdf

Greg F.
Posts: 372
Joined: Wed Nov 23, 2005 6:51 am
Location: Seattle

Post by Greg F. » Thu Dec 10, 2009 10:24 am

"The passive cooling on the CPU does require a system fan, for optimal cooling.
The heat sink measures 36mm off the upper side of the main board.
DO NOT use this board without a fan in the system."
quoted from OP's link. so while it may be a very good solution it is not what I would call passive. It also has D-SUB video and 2 Sata connections.
Last edited by Greg F. on Thu Dec 10, 2009 10:34 am, edited 1 time in total.

xan_user
*Lifetime Patron*
Posts: 2269
Joined: Sun May 21, 2006 9:09 am
Location: Northern California.

Re: D510MO (Mount Olive) : 13watt Atom D510 mITX board

Post by xan_user » Thu Dec 10, 2009 10:29 am

USA=
Price: TBD
Available 01/05/2010


UK=
Shipping from 6th January 2010
Please note, we cannot take pre-orders until it has been released by Intel on approx 4th Jan

pyotr
Posts: 11
Joined: Tue Oct 18, 2005 1:15 pm
Location: the Netherlands

Post by pyotr » Fri Dec 11, 2009 9:51 am

No DVI, no HDMI. :evil:

Crim
Posts: 23
Joined: Tue Nov 03, 2009 2:25 am
Location: Boulder, CO

Post by Crim » Fri Dec 11, 2009 11:51 pm

Any word on a version with DVI or HDMI? Gees, you'd think that in this day and age DVI would be a give in.

aztec
Posts: 443
Joined: Mon Dec 12, 2005 5:01 am
Location: Foster City, CA

Post by aztec » Sun Dec 13, 2009 3:22 pm

I was browsing around and found a couple of early previews that had intially said this board, along with the 410, would actually be closer to the $50 mark than $100.

That would be incredible if it holds true.

Does anyone know how many fan headers this board has? It looks like just 1.

Methanoid
Posts: 8
Joined: Mon Jun 29, 2009 11:44 pm
Location: UK

Post by Methanoid » Wed Dec 16, 2009 2:57 am

Tranquil's price sucks....

Its 70 Euros inc 19% VAT in Germany or £63 (thats with HIGHER VAT)

Its 99 Dollars in USA or equivalent £60

Its 79 Pounds inc 15% VAT in UK - with less VAT..

Hmmm methinks Tranquil are taking a 25% additional profit compared to other firms.

I'm hoping someone in UK is able to source and price a little more fairly. No-one objects to firms making a profit but gouging us is not on!!

Neven
Posts: 1
Joined: Wed Dec 16, 2009 4:06 am
Location: Germany

Post by Neven » Wed Dec 16, 2009 4:10 am

The following hardware was used for our test system:
- D410PT / D510MO
- 2 x 2GB DDR2 DIMM 800 Mhz
- HDD 2,5“ SATA 80GB AC (Seagate)
- Slimline Combo drive CW-8124-B
- Pico 150
- 84 watts AC adapter
An idle power consumption of 26W with this set-up is actually very weak. I have seen quadcore set-ups that consume less power at idle.

anaqer
Posts: 44
Joined: Sun Jun 26, 2005 11:31 pm

Post by anaqer » Sun Dec 20, 2009 11:26 am

Crim wrote:Any word on a version with DVI or HDMI? Gees, you'd think that in this day and age DVI would be a give in.
Wouldn't count on it - Intel imposes certain restrictions on the features of Atom-based designs, and IIRC analogue-only video is one of them.

Did they finally change the packaging of the south bridge, though? Sure looks to be mBGA on that image.

aztec
Posts: 443
Joined: Mon Dec 12, 2005 5:01 am
Location: Foster City, CA

Post by aztec » Mon Dec 21, 2009 3:35 pm

Here's a good review from Anand.

He's also quoting that the D510MO will be $75

http://www.anandtech.com/cpuchipsets/sh ... spx?i=3692

Muse
Posts: 1
Joined: Sat Feb 06, 2010 11:00 am
Location: SF Bay Area

Re: D510MO (Mount Olive) : 13watt Atom D510 mITX board

Post by Muse » Sat Feb 06, 2010 11:12 am

aztec wrote:
For ~$99 and passive at that!

This could be a great platform for a simple WHS box.
I just read a review at Newegg for this mobo (Intel D510MO) by radman on Feb. 4, 2010 (2 days ago now) stating that he communicated with Intel tech support and they confirmed that this and other non-server mobos do not and will not support server OS's. Thus, the video driver support is lacking for WHS and Server 2003. Here's the link:

Link forbidden by SPR

Well, they won't let me post a link since I don't have 3 posts here yet, but I think you can find it.

speedboxx
Posts: 72
Joined: Wed Dec 30, 2009 9:13 am
Location: Canada

Post by speedboxx » Sat Feb 06, 2010 5:27 pm

Neven wrote: An idle power consumption of 26W with this set-up is actually very weak. I have seen quadcore set-ups that consume less power at idle.
Proof/link?

flyingsherpa
*Lifetime Patron*
Posts: 475
Joined: Fri Sep 26, 2003 6:28 pm
Location: CT, USA

Post by flyingsherpa » Sun Feb 07, 2010 9:05 am

speedboxx wrote:
Neven wrote: An idle power consumption of 26W with this set-up is actually very weak. I have seen quadcore set-ups that consume less power at idle.
Proof/link?
Maybe he was referring to electrodacus who has posted a system idle of 31W with a heavily undervolted quadcore (the cpu was using 24W of that).

electrodacus
-- Vendor --
Posts: 372
Joined: Sun Jul 12, 2009 10:30 am
Location: Canada
Contact:

Post by electrodacus » Mon Feb 08, 2010 9:08 pm

flyingsherpa wrote: Maybe he was referring to electrodacus who has posted a system idle of 31W with a heavily undervolted quadcore (the cpu was using 24W of that).
The CPU was using 4W out of total of 31W at idle and the system will use 35W if not undervolted with CPU using 8W at idle.
So if the Atom board is using 26W at idle that is really bad.
Dual core atom D510 get a score of 650 points compared with Q8400S 3600 points according to Passmark list this is more than 5 time less.
The 24W for the Q8400s CPU is at 100% load on all cores at 2GHz and 0.925V and will still score 2700 points in Passmark so it can beat the D510 with only one core :)
They can get much better power consumption on Atom boards but they do not want ( I do not know the reason ) one big problem is that the CPU is the only new technology on the board the rest of are history including integrated video also I think they use much higher voltage than needed just to increase the power consumption.
I have an ITX board that is using an ULV Celeron 1GHz and 5W TDP and this was produced in 2005, also can score 260 points in passmark and is a single core at 1GHz.
You can see more here Link Celeron ULV vs Atom vs Q8400s

HFat
Posts: 1753
Joined: Thu Jul 03, 2008 4:27 am
Location: Switzerland

Post by HFat » Tue Feb 09, 2010 12:22 am

electrodacus wrote:They can get much better power consumption on Atom boards but they do not want ( I do not know the reason ) one big problem is that the CPU is the only new technology on the board the rest of are history including integrated video
They DO get much better power consumption on the boards which are designed for very low power consumption such as the Poulsbo boards, which are of course more expensive and slower.

The point of this board is that the power consumption is low enough for out-of-the-box fanless operation (in theory anyway) while being relatively powerful. This allows one to build really cheap no-nonsense systems. What would be the point of even lower consumption if it's going to be connected to the grid?

Obviously Intel is seeing the potential of this board or they wouldn't have crippled it (no DVI, poor LVDS resolution, slow PCI, no support for MS servers and so on). I doubt they actually went out of their way to increase its power consumption.

The integrated video is integrated to the CPU... so I'd say that part is rather new actually.

wywywywy
Posts: 69
Joined: Tue Sep 16, 2008 11:47 pm
Location: UK

Post by wywywywy » Tue Feb 09, 2010 6:11 am

Grrrr only two SATA ports again :evil:

HFat
Posts: 1753
Joined: Thu Jul 03, 2008 4:27 am
Location: Switzerland

Post by HFat » Fri Feb 12, 2010 9:36 am

I got myself one of these and I posted in the article discussion thread: viewtopic.php?t=56735

I'm posting here as well so as to let everyone interested know I'm taking questions and test requests.

andymcca
Posts: 404
Joined: Mon Jan 11, 2010 8:19 am
Location: Boston, MA, USA

Post by andymcca » Fri Feb 12, 2010 10:10 am

HFat wrote:The point of this board is that the power consumption is low enough for out-of-the-box fanless operation (in theory anyway) while being relatively powerful.
Based on how warm my <15W idle system gets in the M350 case, I cannot imagine it would work out well to have one of these ~30W idle machines in the same case without fans. It would be shocking to me if it did not hit 55C+ idle.

HFat
Posts: 1753
Joined: Thu Jul 03, 2008 4:27 am
Location: Switzerland

Post by HFat » Fri Feb 12, 2010 11:09 am

Why would 55C idle be unacceptable? I certainly wouldn't worry if my laptop's CPU idled at 55C for instance considering how hot many laptop parts get in spite of their fans when you're stressing them. It's the other components which concern me.

HFat
Posts: 1753
Joined: Thu Jul 03, 2008 4:27 am
Location: Switzerland

Post by HFat » Sat Feb 13, 2010 7:18 am

andymcca wrote:It would be shocking to me if it did not hit 55C+ idle.
Read my latest update and be shocked. ;-)
I think one should get better temps with the M350 than with my case but what do I know?

andymcca
Posts: 404
Joined: Mon Jan 11, 2010 8:19 am
Location: Boston, MA, USA

Post by andymcca » Sun Feb 14, 2010 8:00 am

HFat wrote:
andymcca wrote:It would be shocking to me if it did not hit 55C+ idle.
Read my latest update and be shocked. ;-)
I think one should get better temps with the M350 than with my case but what do I know?
I'm now thinking that your case has better coupling directly to the board (doesn't it have some thermal pad under the processor/NB?). That may explain the better temperatures, since the M350 relies almost totally on convection on the tiny CPU/NB heat sinks.
55C idle only worries me because of what it implies under load :) The N270 has a TJmax of 90C, but I don't like to see temperatures anywhere near that since the temperature may not be accurate or represent the hottest part of the chip. And I have no clue how hot the NB is getting, but since that is the 20W component in a 330+945GC combo, it is probably way hotter than the CPU. (although someone was saying TJmax for 945GC might be 125C? I have not seen this on a data sheet though)

HFat
Posts: 1753
Joined: Thu Jul 03, 2008 4:27 am
Location: Switzerland

Post by HFat » Sun Feb 14, 2010 9:24 am

andymcca wrote:I'm now thinking that your case has better coupling directly to the board (doesn't it have some thermal pad under the processor/NB?).
I don't think so (no thermal pad or anything like that stock). The only way in which it would be superior is that the metal must be spreading the heat a bit. Forgive my ignorance but the M350 is plastic, right? Then again the hottest part of the metal is heated by convection and hot air must be passing through the mesh anyway.
andymcca wrote:That may explain the better temperatures, since the M350 relies almost totally on convection on the tiny CPU/NB heat sinks.
55C idle only worries me because of what it implies under load :)
The heatsink on the D510MO looks a lot bigger than the one on the D945GSEJT (I've only seen it in pictures).
In ideal conditions (see the thread linked above for caveats... and I'll be posting more shortly), the differences between idle and CPU load are relatively small. That's because Atom processors have a very low TDP and also probably (I guess) because they're consuming a significant share of that at idle. I'm not seeing a significant temperature difference between low CPU load and idle (like you'd get with more powerful processors that downclock and downvolt dynamically).
andymcca wrote:The N270 has a TJmax of 90C, but I don't like to see temperatures anywhere near that since the temperature may not be accurate or represent the hottest part of the chip.
As a matter of fact, in the case of the D510MO, lm_sensors picks up additional "core" temperatures that diverge significantly under CPU load (they get hotter that is) from the "processor" temperature reported in the BIOS screen.
andymcca wrote:And I have no clue how hot the NB is getting, but since that is the 20W component in a 330+945GC combo,
The D510's NB is integrated to the dual-core Atom and their combined TDP is 13W.

jessekopelman
Posts: 1406
Joined: Tue Feb 13, 2007 7:28 pm
Location: USA

Post by jessekopelman » Sun Feb 14, 2010 1:52 pm

andymcca wrote: 55C idle only worries me because of what it implies under load :)
One aspect of Atom is that TDP difference between idle and load is very small. So if idle is 55C, I'd be surprised if load is > 65C, with no change in cooling.

HFat
Posts: 1753
Joined: Thu Jul 03, 2008 4:27 am
Location: Switzerland

Post by HFat » Sun Feb 14, 2010 2:18 pm

jessekopelman wrote:So if idle is 55C, I'd be surprised if load is > 65C, with no change in cooling.
It depends on the cooling and on what you're talking about but I figure the "core" temperatures reported by lm_sensors are likely to rise by about 20C between idle and CPU load if the D510MO is laid horizontally without active cooling for instance.

Post Reply