SB750 Southbridge
While technically a new part, the SB750 appears to us to be a point upgrade to the SB700. However, it is a major improvement over the SB600. Anyway, let’s get into what has changed and what has not.
The SB700/750 features six SATA 3.0Gb/s ports, up from four on the SB600, with the ability to reserve up to two of those ports for eSATA connectivity. AHCI performance and setup is much easier now and we have not experienced the same problems that plagued the SB600. Drives can be set up in RAID 0, 1, or 10 and RAID 5 now makes its appearance in the SB750. RAID 5 works well for a host based controller and even outperforms the ICH10R in early testing with a three drive array, but four or more drives is resulting in some odd behavior that we are trying to figure out now.
Still missing in action is a native interface for networking support. AMD continues to use an external PHY and MAC for network operations. Although performance is similar to the NVIDIA and Intel solutions, this setup does incur a cost penalty for the motherboard suppliers.
The major improvement in the SB700/750 series over the SB600 is the increase in USB 2.0 performance and the number of ports available. The new dual-channel controller features 12 USB 2.0 capable ports and 2 specific 1.1 ports for compatibility reasons. USB 2.0 performance is up to par with the Intel and NVIDIA solutions.
A single PATA channel provides native IDE support for up to two drives. This channel supports PIO, multi-word DMA, and Ultra DMA 33/66/100/133. Six PCI lanes are still included although we doubt a board manufacturer is going to offer that many. AMD dropped HyperFlash support on the SB700 to make way for the new Advanced Clock Calibration interface.
The SB700/SB750 features four PCI Express lanes for the A-Link Express II interconnect between the Northbridge and Southbridge, but like the current 790FX/780G series, those four lanes are based on PCI Express 1.1 specifications. That means the interconnect bandwidth is capped at 2GB/s, half of what it would be in a PCI Express 2.0 configuration.
Finally, we have the High Definition Audio controller carried over from the SB700 that allows up to 16 channels of audio output per stream. The controller supports up to four codecs with sample rates reaching 192kHz at up to 32-bits per sample.
ACC-
The most drastic and perhaps most important change on the SB750 is the new Advanced Clock Calibration interface that promises additional overclocking headroom or lower voltages with the Phenom processor series, especially the BE products. We have seen clock improvements ranging from 100MHz to over 500MHz depending on the processor we utilized.
The SB750 now has a direct 6-pin interface to the AM2+ socket on the motherboard, there are now pins on the Phenom CPU that connect directly to the SB750. These pins were previously unused and are now used as a means of communication between the South Bridge and the CPU. The SB750, in combination with an updated BIOS, can now override some of the CPU's internal settings which can potentially increase the overclocking headroom of the chip.
AMD says that the settings tweak doesn't impact performance and doesn't change thermals or voltages, it simply can allow a Phenom processor to clock higher when overclocking. The BIOS exposes the parameter being changed, which AMD refers to as the Advanced Clock Calibration (ACC) value. Typically this value has a range of -2 to 0, on motherboards with the SB750 that support ACC the value can be set from -12 to +12. Higher numbers should allow for higher clock speeds, while lower values should allow for lower voltages/lower power operation.
Even after a full press court for the last three weeks, AMD will not tell us what this value actually adjusts - simply stating that it makes it easier for the CPU to run at higher speeds. Based on AMD's careful choice of words, it would seem that adjusting the ACC somehow changes the acceptable margins of operation for the CPU cores (the value can be changed on a per-core basis in AOD or BIOS). By loosening these margins, however it is able to do so; the SB750 + ACC combo can enable many Phenom processors to operate outside of their normal overclocking margins. In fact, AMD is confident enough about the technology that they are willing to state publicly that on average; a 100MHz to 300MHz increase in clock speeds is attainable by the user after some tweaking. In fact, we totally agree with these numbers after extensive testing.
The SB750/ACC trick doesn't work on anything other than Phenom processors, and it works particularly well on the Black Edition processors. While AMD didn't rule out eventually enabling this on K8 based cores, it is a Phenom-only option for now. Reaching higher clock speeds is more of a top priority for Phenom, and it's unclear whether or not whatever AMD is doing here can even work on K8 if they tried.
AMD also committed to enabling similar tweaks for upcoming 45nm based Phenom parts, implying that this was not a short-term solution to the clock speed problem. At the same time, AMD implied that the tweaks that the SB750/ACC feature enables could be incorporated into the manufacturing chain and actually implemented in hardware.
Microprocessors are designed to operate in even the most extreme of conditions, AMD seemed to imply that its ability to adjust the ACC value somehow changes this. Curiously enough, AMD cited "competitive concerns" as a reason why it would not disclose exactly what's going on with this new overclocking feature. We still cannot help but wonder if it is because AMD is going a little too far in the sacrifices, it is willing to make in the quest for higher clock speeds.
Facebook
Twitter
RSS
39 Comments
View All Comments
mpjesse - Friday, August 22, 2008 - link
Did the writer get sick? Did he DIE? 2+ weeks to benchmark an integrated chipset? Practically the entire galaxy has already reviewed this chipset. I'm starting to get tired of both the physical speed (or lack thereof) of this website and the quickness (again, lack thereof) in which crap is published.-Fed Up
bravomail - Wednesday, August 20, 2008 - link
Here is the article with the pictures (in russian unfortunately)http://www.riddik.net/publ/1-1-1-25">http://www.riddik.net/publ/1-1-1-25
The guy ripped apart AMD BE CPU only to find out some bubbled thermopaste between chip and attached lid.
Use Google translate to read this story.
computerfarmer - Tuesday, August 19, 2008 - link
It is now the 19th of August, perhaps the review is coming soon?Beta BIOS was on the 13th of August.
Waiting
Waiting
Waiting...................
OCedHrt - Thursday, August 14, 2008 - link
With all this advancement in IGP, why does Hybrid Crossfire only work with low end cards? I would love to be able to use it with a 4850 or 4870 where the IGP switches completely over to the discrete as needed.MikeODanyurs - Sunday, August 17, 2008 - link
The new DFI 790GX card states it has Hybrid CrossFire that will work with HD 3850 and I believe will even work with the IGP with 2 cards.someguy11 - Wednesday, August 13, 2008 - link
"unless AMD can implement something akin to HybridPower from NVIDIA, which allows the discrete card to power down and switch to the IG unit at idle for less demanding tasks like Office Applications or Bejeweled. Otherwise, the IG unit and all of its potential benefits are wasted with a discrete card setup"Does anyone know if the implementation of something like HybridPower from AMD on the current day 790GX boards is possible?
In other words, if I buy a 790GX will a power saving feature where the discrete card can be powered down be simply a BIOS flash or will new boards have to be made (assuming AMD develops something along this route)?
I ask b/c this "power down" feature really interests me and if this can be implemented via a BIOS upgrade, then I am more apt to get a 790GX today rather than a 790FX board who knows when. I also like the idea of having a board with onboard IG so you can play the graphics card price wars when you want to.
Thanks for the help!
ZootyGray - Tuesday, August 12, 2008 - link
There are lots of unbiased websites offering useful reviews. I am sick of waiting for your uncompleted works and biased results.Your review of 4870x2 is the end of the line - that is the worst trash I have seen.
Why would anyone want to wait for you to trash more AMD hardware.
Save your energy for yourself. I do not need your self-importance.
Maybe when you can followup on last May's 780G review, as promised, I might be interested - until then - I apologize for thinking you were some kinda good review place. Yer out. Bias is not quality. It's abuse. Not your victim anymore.
computerfarmer - Saturday, August 9, 2008 - link
On page 2 it is noted as saying the new 790GX graphic is set(default) at 700mhz compared to the 780G at 500mhz. The manual for the Gigabyte GA-MA790GP-DS4H (rev. 1.0) page 42 has the default set at 650mhz. This leads me to believe that side by site tests from different manufacturers should show differences based on default settings. Also SidePort memory ratings need some attention. If SidePort is DDR3 at 1333mhz and system memory is at 800/1066mhz then how do they function together at different speeds? Or is the SidePort memory speed based on system memory up to 1333mhz?Thank you Anandtech for such an informative article.
maree - Friday, August 8, 2008 - link
Quoting from the article"For the multimedia user, ... The SB750 does offer ACC and along with it the opportunity to run lower voltages ..."
Does it mean motherboards running on this chipset can take a phenom 9550/9850BE an turn off 2 cores then expect the power performance of amd 4850e?
ZootyGray - Thursday, August 7, 2008 - link
I really don't want to go dig up exact quotes from previous reviews."The Three Amigos" review revealed "pop goes the mosfet" issues with products produced by various mobo manufacturers. I really liked that review - BUT - a followup was indicated as coming - and it never appeared.
This review ends in a similar manner.
And the Foxconn review seemed to really sidetrack and resulted in 3 articles; and for my little brain, I found it difficult to follow the coherence of the whole thing.
Now we have a whole new chipset and a bunch of mobos you say you are benching soon. Ok, so that makes sense - assuming this will actually happen soon. No problem there.
BUT - there's this really interesting comparison to 780G which you are presenting. I wonder what happened to the mosfet issue that you so beautifully revealed. I wonder - Is it in play here as well? I DOUBT that! This is a performance board - right?
BUT - are you trying to lead some people to buy the original 780G boards regardless of resolution (if any) of the "pop goes the mosfet" issues. And asus has produced new mobos that clain 125watt handling power - ie no "pop"???
I think all this uncertainty might be dangerous for some people with great expectations.
I think some unfinished business should be clarified.
I think some comment on 'mosfet' should be included in these upcoming benchies.
I think this might be misleading a lot people - and your unfinished work is a little underhanded slight on AMD products. I REALLY HOPE you clear this mess up; since it was your baby to start with - and that was good. And that caused me to respect Anandtech.
Please don't leave me just hanging on and waiting - I doubt I am the only one. I would prefer completion rather than assuming this is another antiAMD site, and have to go elsewhere. It's your credibility, for me, at least, at this point.
Golf taught me that follow through is a big part of the golf swing. I know you are busy. Some loose ends can bite you. How close IS this to the 780G boards anyway? Has the "pop" been fixed?
I hear Vanilla Fudge "Keep Me Hanging On" playing in the background - wow, it has been a while since I heard that.
Thank you.