Intel P35: Intel's Mainstream Chipset Grows Up
by Gary Key & Wesley Fink on May 21, 2007 3:45 PM EST- Posted in
- CPUs
Final Words
The P965 launch is still too fresh in our memory to ignore. It appeared the roughest launch in recent Intel memory with a BIOS a day, and sometimes several a day, in the early going. Many of the features did not work, or at least did not work consistently until several months of BIOS updates. While the hard work has paid off for manufacturers and P965 is now a capable chipset, it is still not the paragon of reliability that we normally expect of Intel chipsets.Contrast that to today's performance launch of P35, and the differences are striking. P35 is very mature at launch and is generally a joy to work with. That is not to say it is perfect, as manufacturers are still scrambling to fully support all the new added features of the chipset. The base performance, however, is stable, and P35 returns us to the confidence we normally feel for Intel chipsets.
Some will say this is the result of the fact that P35 is just a mild upgrade of P965, fixing what is wrong with that chipset. We don't agree, since features like 1333 processor bus and support for DDR3 memory to 1333 speed and beyond are significant additions that definitely impact performance. Also significant is full support for the upcoming 45nm Penryn processors. As you have seen in the P35 benchmarks these are not just minor "fixes" - they are features that have a significant impact in improving system performance. No doubt Intel did tweak the base P965 code in the process, and we are grateful for that, but P35 introduces too much new that improves performance to just be considered a respin.
Looking at the four motherboards evaluated in this launch review, it is clear that all of the boards are very good with existing features like memory to DDR2-1066 and the current 1066 processor bus. However, when we move into the new performance areas of 1333 processor bus, DDR3, and 1333 memory speed, the ASUS P5K3 and P5K twins have the edge over the Gigabyte and MSI offerings. The MSI and Gigabyte are very good boards, and BIOS updates received during testing have us expecting them to be the equal of the ASUS boards in another BIOS spin or two. In fact, the Gigabyte BIOS received this weekend moves the P35 DQ6 into performance parity with the ASUS boards in early testing. For now, however, the ASUS pair is more polished in their implementation.
You will not, however, have to suffer with the Gigabyte or MSI as they both perform exceptionally well with the components we now have. Any of the four platforms will provide a rock solid base for a new system, performing better than just about any P965 board we have tested. You will not have to struggle with these P35 boards as many of you did with early P965. We also fully expect the Gigabyte and MSI to mature quickly in the "added features and refinement" category of P35.
What we did find in testing the P35 boards is that BIOS tuning is paramount to extracting the best possible performance from each board. ASUS's BIOS is very mature at this point and the extra features like Transaction Booster and CPU Voltage Damper actually make a difference in the performance of the board. We have also noticed the manufacturers starting to open up the BIOS with settings that will greatly assist the enthusiast in extracting the best possible performance from this chipset. However, we would still like to see the ability to change chipset straps and additional memory settings in the high-end P35 products.
Overall performance of the P35 chipset was impressive with it constantly finishing at the top of our benchmarks. Even though the differences in scores were minimal at times, the level of consistency shown by this chipset was impressive. The addition of DDR3 memory support, the improved memory controller, and the upcoming 1333FSB processors indicate that this mainstream chipset is a force to be reckoned with both with today's components and those on the horizon.
In current games like Prey we found a 9% increase in frame rates between the ASUS P5K3 at 1333/1333 and the Gigabyte P35-DQ6 at 1066/1066. Granted, the move to a 1333 FSB provided about a 6% improvement in frame rates on the DDR2 platform. However, once lower latency DDR3-1333 is available we expect the combination of this memory and the switch to 1333FSB processors to possibly create double-digit improvements in applications that are memory bandwidth and latency sensitive. Our main concern at this time is the cost and lack of widespread availability for DDR3 memory. Latency and speed improvements are developing quickly for DDR3 but cost appears to be the biggest stumbling block for entry into this technology now. Hopefully this will not be an issue six months from now as DDR3 ramps up in both production and system installs at major OEMs.
The ICH9R brings improved disk performance and in early RAID 5 and 10 testing we are seeing a 2% ~ 4% improvement in certain applications when compared to the ICH8R. The ICH9 also blesses us with an additional two USB ports and improved performance. Typical benchmarks like HD Tach or HD Tune do not show a difference in USB performance over the ICH8R currently; however, we are seeing measurable improvements in write and read speeds under Vista when comparing the two chipsets and a few minor compatibility issues seemed to have been solved. While not up to the speed of NVIDIA's USB solution, it is nice to see Intel making incremental improvements quickly.
We have explained the benefits of the chipset and provided an early look at its performance potential. What are the drawbacks? The P35 performance is impressive and the chipset along with the first wave of boards seem to be very solid. The problem is the cost of entry might be more than one is willing to spend to upgrade from a recently purchased P965, 680i, 650i, or even 975X board. Power consumption was a concern for us. Based on the current chipset TDP ratings we expected the power envelope to be near the P965 or possibly improved depending upon how aggressive Intel implemented power saving features on this chipset. Instead, we end up with a chipset that now consumes more power than the 680i and requires even more elaborate cooling. Something the motherboard companies have obviously taken to heart by the amount of heat pipes on the boards now. We doubt all of it is really needed but it is amusing for the time being to see the manufacturers trying to outdo each other in both design and amount of aluminum they can squeeze into a three inch square space.
There is no doubt after testing P35 that it is clearly the best-performing chipset you can buy today. Intel has done an excellent job in the added features, and performance is superb. P35 is at its best in gaming and memory-intensive applications, where the performance improvements are most dramatic. If you are building a new system there is a lot to like with P35, and it all starts with improved performance.
Facebook
Twitter
RSS
58 Comments
View All Comments
drewintheav - Sunday, May 27, 2007 - link
When you tested the P5K Deluxe Wifi mainboard and OCZ Reaper 9200 ddr2 memory were you able to boot up with with both modules of the reaper memory installed or did you have to boot up with a different memory and then adjust the ddr2 voltage to 2.3?I can not boot with all settings on the Asus P5K Deluxe Wifi set to default. The motherboard starts up but the screen stays black and there is no beep or POST information displayed.
Were you able to boot up no problem? I was able to boot no problem at default settings on my Asus Commando with the same OCZ Reaper 9200. Do you think I need to replace my motherboard? I also have had the pc spontaneously restart twice now.
Thanks in advance for any info or suggestions that you can provide.
Stele - Tuesday, May 22, 2007 - link
Most sites seem to list Asus's high-end boards as having 8-phase power, as per the spec sheet provided by Asus. However, I've a funny feeling that it might not be so, despite the number of MOSFET "sets" and inductors. When I was working on P5B Deluxe boards, which also supposedly sports 8-phase power, I noticed that they used the Analog Devices ADP3198 synchronous buck controller. This controller supports up to 4 phases max, which suggests that the 8 apparent phases are actually arranged in two sets of 4-phase circuitry running in parallel. If this is true, it's an unfortunate bit of misleading marketing, though it was probably due to the pressure in keeping up with gimmicky competitors touting quantity (6-phase, 12-phase, xx-phase) over quality (component/circuitry design/implementation).The presence of a 'voltage damper' setting in BIOS for the P5Ks seem to hint at the possibility of something new, however: either the presence a new PWM controller or some additional droop compensation circuitry. I'm guessing it's still an ADP3198, and hence the latter scenario. Would it be possible to find out what controller's being used on the P5K, pretty please? :P
IMHO, while the Analog PWM controllers have generally been stellar performers in their own right, they ought to start improving their products so that manufacturers would have to worry less about breaking relationships with them and jump ship - especially with, for example, Intersil having excellent analog (up to 6 phases) and digital PWM controllers.
Stele - Wednesday, May 23, 2007 - link
Now that I recall, Gigabyte too is in the habit of being rather liberal with how they count phases... in fact they're arguably one of the first to do so, along with their DPS gimmick in the 875 days. The DQ6 is listed and marketed has having 12 phases, but if I remember correctly, Gigabyte uses an Intersil ISL6327 for their 12-phase products (at least the GA-965P-DQ6 did so). This is a 6-phase controller, which means that, as in the case of the Asus designs mentioned above, the "12-phase" power claimed by Gigabyte is little more than two sets of 6 phases in parallel. Is this still the case in the P35-DQ6?If so, it'd be good to point out the reality under misleading hype, so that if nothing else, readers and potential buyers are better informed - and to underscore the fact that there are hardware sites who do know their stuff and who don't simply swallow and parrot whatever the manufacturers throw at them.
Just my 2 cents'!
shabby - Monday, May 21, 2007 - link
The loop-de-loop kicks ass on the msi board :DTA152H - Monday, May 21, 2007 - link
I agree, I don't know why, but I need it!Gary Key - Tuesday, May 22, 2007 - link
It actually works also, a lot less aluminum than the Gigabyte and ASUS solutions, yet the MCH/PWM area seemed to be within 1C of the other boards in our early testing with the Quad. Still not a big fan of the "designs" coming out but to each his own. I think the new abit P35 board as the best looking heatpipe system but then I like it old school... :)TA152H - Wednesday, May 23, 2007 - link
I don't like you any more. I'm just not sure I can relate to someone that didn't like that rollercoaster :P .I expect it probably should work pretty well. With windmills, one of the biggest considerations for the power it produces is how tall the tower is; the further from the Earth, the more wind energy it will get. So, by extension, by creating this roller coast they elevate it to a higher point, and of course by convection the hottest air will hit that spot. So, it should work out pretty well (by now you're probably scratching your head thinking "what is this idiot talking about". It's because I have no clue, but I want it, and I'm making stuff up to validate the decision like this windmill nonsense).
Old school heat pipe systems? Hmmmm, I don't remember even seeing a heatsink on anything earlier than a 486 processor, so I just think we can categorize your heatsink taste, at best as prosaic, and maybe even boring! I think this roller coaster only portends of things to come, I expect bigger ones soon, as well a hot dog stand. Motherboards seem to have become amusement parks, we have carousels (spinning fans), water equipment, towers, and now a roller coaster. Good grief.
yyrkoon - Thursday, May 24, 2007 - link
Oh, and . . . .http://www2.abit.com.tw/page/en/motherboard/mother...">http://www2.abit.com.tw/page/en/motherb...pMODEL_N...
yyrkoon - Thursday, May 24, 2007 - link
I agree with Gary, these 'windmill' and butt ugly other heatpipes from other OEMs can not touch ABITs OTES in looks. I guess that is what hapens when you pioneer a technology, and everyone copies . . .Stele - Monday, May 21, 2007 - link
While the addition of 2 more USB 2.0 ports is certainly one of ICH9's new features, doesn't ICH8 already offer 6 SATA ports? That was one of the latter's welcome improvements over the ICH7.
There also doesn't seem to be anything in the block diagrams that indicate the presence of an eSATA port. I'm guessing here, but could it be that the mention of a "port multiplier" led to some confusion? After all, the 6 SATA ports in ICH8 and 9 are actually controlled by two host controllers, with internal port multipliers that connect them to 4 and 2 ports respectively (that's why, for example, some board manufacturers label two of the SATA ports as 'secondary' and state that it is generally not advisable to connect the boot drive to them).
However, I'm guessing that the port multiplier issue came up in the context of ICH9 because apparently Intel would be using only command-based switching in ICH9's port multipliers, removing FIS (Frame Information Structure)-based switching apparently used in ICH8 (most port multipliers use both command-based and FIS-based). While FIS-based switching offers higher performance under multiple-drive loads than command-based switching, FIS-based switching is more complex and hence expensive to implement; besides that, it uses slightly more CPU resources to manage the increased command and data flow rate. As such, Intel might have chosen to use the simpler, cheaper command-based switching for ICH9. If this is true, it might be to keep costs down and because many hardware sites probably over-emphasised on CPU-utilisation when comparing SATA performance with other chipsets. Argh. Again, this was something that came up on the grapevine but I've not been able to find official sources that verify or deny this... perhaps Anandtech could shed some light?
In the event that it's true, exactly how much performance would suffer under multiple-drive read/write loads is unclear, and this is where an extensive non-RAID and RAID performance test with perhaps a full six drives (or even 3/4/5) might give some hint. On the other hand, it may be that performance would not differ much in scenarios short of an array larger than a two-drive RAID 0/1 (as demonstrated in this review), so Intel decided to dispense with what they may consider a needless luxury.