Cool

The MSI board is interesting

RE: MSI Product Uses Less Power

IC's come in a lot of specification combinations. Perhaps the chip that ASUS is using consumes more power for the same speed and costs 1 dollar less. Or, perhaps the MSI engineers were able to design better pcb interconnects and were able to run the NB at 0.1V less.

Hi Mike your comments on wasted time and effort in testing early production models made me wonder why do motherboards in SPCR at all .
I know I will probably be flamed for even questioning this but as I see it the Mobo doesn't have a great bearing on silence. The things which bolt to the Mobo are the things which produce the noise like fans and hard drives or suppress noise like cases and suspensions.
I do not wish to be disrespectful but other specialized sites are probably
better equipped to test motherboards.

busso

The northbridge and southbridge produce heat... heat is the enemy...

Also some of us at SPCR build ultra low power passive systems and enjoy being updated on new products

Most reviews on other sites, maybe even all reviews on all other sites, fail to mention things that matter to us like SpeedFan compatibility. It was good to know the MSI board is currently not supported. I would want to know if/when it is supported, because that's important to me.

+1

Also it would be nice if that board could be tested with the newly available Speedfan 4.39.

Umm, that's not really true. You can't extrapolate anything if you use a 4850e, that completely changes the picture since the bandwidth available to the IGP will be halved. Pretty much any other review has found the graphics core iis quite similar in performance to HD3200, IF the boards are otherwise similar (so, both or none have sideport memory running at the same frequency).

btw can you now use dvi/hdmi simultaneously? I fear that's still not possible.

Thanks, Lawrence!

MSI's board is a really good effort on their part, considering some of the rubbish they have come up with in the past. I like the carton design too: business like, simplistic and doing away with the gaudy and ghastly fantastical and mythological imagery that pervades so much of computer gaming focused marketing. It looks like they have taken a leaf out of Gigabyte's book, even to outdo them on the number and diversity of back panel IO connectors and to the board layout itself. Let's hope these trends continue and that all motherboard makers follow so that consumers get better value and quality for their money.

As the article implies, 785G is an improvement; a continuation of the former --- not a landmark chipset for desktop computing. Perhaps that will be reserved for the 800G, or what ever nomenclature AMD decides to use next time around.

I disagree with a previous poster who said that SPCR may not be best suited to review motherboards. On the contrary, I think SPCR is well suited to do so. While they do not go into all the superfluous time and space wasting of 30 pages of macro photography, synthetic benchmarks and egotistical juvenile overclocking exploits, they provide enough relevant detail to those people who are interested in such things as power consumption, component efficiency, fan and voltage controls. Though largely unfashionable in technology realms, these are the sorts of things,-- after design and feature implementation aspects are considered--that influence many SPCRians' purchasing decisions. Moreover, I think many have become accustomed to Lawrence Lee and Mike C's articles over the years. The consistency from a technical writing perspective of their reviews and the lack of time wasting is what interests me heretofore and why I continue to read the articles.

Avoiding ZOMG!!1! motherboards isn't enough to ensure low power consumption.

The comparison I would like to see is how does the AMD 785G boards compare to the Intel/Geforce 9400 boards like the GIGABYTE GA-E7AUM-DS2H for an HTPC? These are the two boards I am considering. Which board would you choose for an HTPC?

What I am trying to say is that one or two motherboard reviews is not good enough to be making an informed decision on a new motherboard. When I replace my motherboard, not as frequently as I would like ,but about every 18 months to 2 years my first requirement is that it be stable and have all the features for the peripherals I have.
Power consumption and component efficiency is not something most people would consider as a requirement of purchase and speedfan is not the be all and end all of fan control there are many independent fan controllers which do just as good a job and you get that info here at SPCR. There have been about 6 mobo reviews done on SPCR over the years and all different types and I suggest the list of Cool and Quiet Athlon mobo's found on SPCR which uses other sites as a reference is the way to go. By the way there is no up to date list of "cool and quite" motherboards in SPCR and I am not suggesting there be one but it is unfair on all the other motherboards out there for SPCR to select one or two a year to review . That's is of course if there are no other fans, cases, hard drives, coolers, suspensions, power supplies and the like to test. Even more video cards would be preferable. SPCR does what it does better than any one else in the field of silent computing but unless it does motherboards more comprehensibly than 1 or 2 a year results are not the basis for Mobo selection. Interesting as they may be.
And again I say this with no disrespect intended.

bussoguy --



Wha???

You should do the due diligence and research the site before criticizing. I count 17 mobo reviews over the past 12 mos. I brought Larry on board back in 2007 specifically for mobo reviews. I haven't counted how many he's done since the start, probably avg 1~2 a month. For the most part, we've focused on smaller (mATX and mini-ITX) boards that get short thrift on most other sites, with an emphasis on usability, onboard video capability for multimedia, fan control, and power efficiency.

No one does any real stability/reliability testing, that requires many samples of each board and long term testing. What most site do focus on is how overclockable a board is -- with 20 pages of benchmarks. This does not give much useful information; the difference in performance between boards is usually paltry.

There's no need for any updated CnQ mobo listing -- that's defacto on every AMD board, it's the odd one that doesn't work which is the anomaly.

Exactly. That's why I try to read as many reviews as possible for each alternative board I'm contemplating before making a decision.

SPCR provide some information that most other reviewers either totally neglect or just do a sloppy job with. (Applies to all types of hardware.)

Cheers
Olle

OK sorry I raised it. I knew it was a bad move as I couldn't express what I wanted to convey.
Will crawl back into my shell and lurk.

You expressed a view and it wasn't agreed with. It happens all the time. No need to feel tortoisey about it. Yet I do see where you're coming from. But that's why you should visit several sites to get an overall picture. Despite my earlier (tongue in cheek) criticisms of other hardware sites' motherboard reviews, there are plenty of them who cover the essential aspects quite well. By all means, read those reviews, then come back to SPCR and get the information that other reviewers overlook to get an all around picture. Plenty of people who buy computer parts aren't all that interested in overclocking and tweaking every one of the ten thousand options in the BIOS. Perhaps that's where SPCR fills part of that void for the less fashionable aspects of hardware as I mentioned in my previous post. They certainly have the right equipment to do the testing reliably.

To be fair to MikeC and Lawrence Lee, there has been a greater focus on motherboards at SPCR in the past year or so. The expectations for a good motherboard and the fierce competition between the main manufacturers means that they are going to be an important talking point when discussing PC hardware for a long while to come.

This is a great site overall. I hope you can glean more out of it, whether you choose to lurk or actively contribute.

Being someone who watercools his machines, I always look at the placement of the chipset heatsink that is below the expansion slots.

Looking at the MSI, the heatsink is directly below the PCIe slot and may prevent someone from watercolling the chipset and having a discreet video card instaled. With a heatpipe connecting the power modulation are to the north bridge (is it a nb?) both units must be cooled if a replacement cooling is wanted.

As for the Asus, the heatsink is directly adjacent to the primary PCIe slot. The second PCIe slot is far enough not to be of concern. The distance of the heatsink is more towards the bottom of the board and the mount point is at the bottom. This may allow the heatsink to be watercooled and not interfere with a discreet video card. The down side is the I/O ports are very close to the heatsink and could have some clearance issues with some coolers. This board does not have a heatpipe which allows the north bridge(is it a nb?) between the PCIe and CPU to be cooled independently of the power modulation area.

thinking of the same thing myself. I have concluded that video performance is good enough on either chipset. The 9300 uses a little less power, but costs about twice as much.
I have an Nividia 8200 that works very well, except that the picture stutters when asked to play 1080i OTA. I guess this is a well known problem when you have a processor with only a 2000mhz hyper transport on this chipset. The solution is to buy a cpu with a 3600HT. I wonder if I could just buy a new 785 motherboard and use the older Brisbane cpu. If it would play 1080i and properly stream flash this is what I would do. But someday I need to buy a newer cpu anyway.

I have the Gigabyte GA-MA785GMT-UD2H board with a Regor 245. Here's the BIOS options...



I'm really liking this thing. Since the voltage adjustments are relative, you can undervolt and keep C'n'Q working without the need for software solutions. I'm undervolting by -0.2V which gives me 0.8V idle and 1.15V load. I also found that AMD C1E was disabled in the BIOS by default. Turning this on shaved another 3W off idle voltages. That with the undervolt got me an idle power of 34W. But similar to my other Gigabyte board, it overvolts the memory by 0.1V with no option to lower it. My 1.5V DD3 memory is reported to be running at 1.6V.

The fan control doesn't have any real options though. Just Smart Fan on or off. Speed fan displays plenty of options but in my minor attempts to adjust things, the fan speeds never changed. Using Gigabytes EasyTune6 I was able to adjust the CPU fan control. You can change the temperature range and min fan speed but only based on the CPU temp. And the GUI looks way better than ET5.

Sorry to have upset you Mike but just a final note. There are the 17 or so boards in the CPU motherboard section.. Yes I can't count and I proof read crap. I did only count 16 but only one or at most 2 of any one type of board chip-set has been reviewed over the last year eg An Asus PQ5 done in Nov 2008 is the only review with that chip-set. I should not have gone away from my initial thoughts.

I go to SPCR every day, just like getting my email. I have gained so much from SPCR. I am currently building a new computer and the CPU HS and fan, the new hard drives, the PS, the case, fan control and noise abating materials have been already purchased directly from the recommended lists or forum info in SPCR . I don’t yet have a EX58 Mobo to fit an intel i920 to.

I still maintain that the Mobo has no noise or noise the ear cannot detect. Of course there will be components ,particularly chokes, which will hum but these will be defective components. But if SPCR did a review of one or even 2 EX58 Mobos there would be insufficient comparison with all EX58 boards to make an informed decision. There are 5 major companies producing EX58 boards each with at least 2 models (Asus and Gbt have least 5) and taking the intel 775 my local shop has 7 companies providing over 35 motherboards for current CPUs. Similarly my local has 25 AM2 and AM3 boards (total) for current AMD CPUs. I would imagine that a similar number of new models will appear before the end of the year. Then there are servers and set top box boards What I am trying to say is that there are a lot of boards out there and for SCPR to make a meaningful contribution to selection it will have to provide the mass of reviews for each chipset similar to the lists it provides for all the things which bolt to the Mobo.

I haven’t said the reviews have not been professional, they are and I have enjoyed reading them and provide they another angle but when the Mobo has little to no contribution to the noise a computer generates why put time and money into the review of one or two chipset models. Sort of do many or none.

I do not wish to disrespect you Mike as its your show and you carry the buck and have more background into the business than I.
There shall be no further posts to this thread by me, promise.

For Shamgar: Re reading my last post it was more turtley than it should. Yes no one supports my idea so its dead but SPCR is too important to me not to be there every or most days.

BillBuerger, I hope Gigabyte and others can improve their bundled software to give more useful options for fan control and power saving features. Gimmicky overclocking options with car racing and secret agency connotations sells to the population who are drawn in by such, but both sides of the equation should be catered for evenly IMO.

bussoguy, I hope you weren't offended by the tortoisey remark. My sense of humour does get the better of me at times. I still don't agree with you on the point of motherboard reviews at SPCR. However, I appreciate (and I'm sure others do also) you explaining your point of view.

You must remember that SPCR has a different focus than most other hardware sites so reviewing lots of server and high end gaming boards would not be profitable (in more than one sense of the word) nor a good use of resources. Because it is not a "big" site either (when compared to other well known hardware sites), companies don't send in the amount of gear to SPCR as they would the others.

MikeC has tried to be more diversified in recent times by reviewing portable gadgets, monitors and other peripherals. Yet that wasn't met with much enthusiasm by forum members,--even to fierce criticism, which I personally think is grossly unfair and unwarranted. Even though this is Silent PC Review, it still has to do with Personal Computing after all and many electronic devices these days are interrelated and cannot function without the use of a PC.

I too, have benefited a lot from SPCR and visit it as much as my time allows me. Some of the products reviewed are not available here but things are much better than they used to be.

Where as ET6 does have some gimmicky stuff going on, my point of posting it was to show the fan control part which overall isn't bad. And this board has really good power saving features as noted by my 34W idle measurement. And it does that without the need for any software which is even better in my opinion.

Very interesting, I really like that approach. Thanks for posting.

Okay. Thanks for the perspective. While I did say "gimmicky", I realise those performance features can also be useful. Bundled software has also improved from what it used to be. Let's hope your good experience with this platform is indicative of the wider experience of others. It's a step in the right direction.

Any info yet on the asus power consumption with latest bios?

As usual, good job! Well done! Thanks for all of yours behind the SPCR curtain.

Though off topic, I'd like to suggest some food for thought: to verify BETTER implementation of Speedstep and CnQ on both side.

Yes, nowadays most if not all motherboards have Speedstep/CnQ functioned correctly under default setting. However, when you tinker system with something exotic like OC/OverVoltage/UnderVoltage, Speedstep/CnQ may soon fail to function at all. Would you mind testing this for SPCR readers? I ask this since I believe SPCR is the sole one who would like to bother with this subtle variance. Anyway, thank you for your hard work.

Btw, based on my experience, AMD boards tend to have BETTER CnQ implementation compared to Intel side.

I've been wondering about building a Phenom 965 mATX-setup, but we'll have to see about that one. i7/mATX is gets me even more worked up.

I found out during my last build (780G) that the boards that I looked at (after discovering the one that I purchased had this problem) had limited resolution when equipped with the HDMI port. I have a 30 inch monitor so this is a big deal for me. I finally found a board that supported the native resolution of my big monitor. It seems the presence of the HDMI port limits the DVI to the max HDMI resolution on the 780G chipset, has anything changed with the 785G?
Steve

Edit - I now see that the Asus board does support 2560X1600 via the DVI port. Still no info on the MSI.

Srbliss, only dual-link DVI or DisplayPort can address the full resolution of your 30incher. I don't think this has anything to do with the GPU.