Rattler unboxing - XS Exclusive

[PiLsY]

There will be a full preview post going up in the reviews section, however that will take a few days to compile as i'm doing lots of testing as I go.


Note: This is a preview board and is NOT final retail. There may be some changes between this board and the retail board but I would expect them to be minor if any are needed at all.


This post will grow as I progress through the review - you'll see all the pictures here as I progress with the board. It is a pre-retail board and im essentially bug hunting - normally this sort of thing wouldnt be public. However, as a treat for you guys Carl has given his OK for me to publish things as I go - no NDA. I'll be trying to show you the features of the board and its general abilities, but bear in mind development is still in progress.


A MASSIVE thanks to Carl at Foxconn UK for supplying the board and for allowing this to be documented publically.


With that said below is a selection of pics. The full res album can be viewed here. Hope you all enjoy the show!







































V2.0 before launch - All rattler boards will be v2.0+ and carry the B3 chipset.





More to come as testing progresses.

[Dualist]

Brilliant, been waiting for this to show up.
Love Foxconn boards but never liked the way they handle unlocked multipliers.
Look forward to your results.

[Lord Trident]

what the problem with unlocked cpu do you mean
is this with QF side or the normal foxconn side

[G.Foyle]

Looks nice and well thought out*, I hope it works as good as it looks

* - except voltage measuring points, but no one ever got them right

[PiLsY]

Spent last night chasing down bios bugs and sorting drivers, made really good progress .

Bios screenshots will be up tonight.

I will tell you that this board kicks ass with Hyper based ram. Missing a couple of ram timings at the moment which i've requested be added (notably RTL and Command Rate), but ive already maxed the board @ 109bclk 21.33x divider @ 1.598vddr cas 8 using a 4gb Patriot Sector 5 2250mhz 9-9-9-27 kit. No stability testing yet and cant do cpuz on this bios, but its in windows, linx stable and "feels" stable. Going to do some tight timings tonight, then BBSE, Micron and Samsung testing to follow. Considering the bios is in an alpha state the memory clocking is very impressive.

[PiLsY]

UEFI bios shots - sadly no windows interface but the support is there to add it later. Aegis clings on a little longer!





















































And the bit we all want to see....

[Mescalamba]

Out of curiosity, how fast UEFI on Rattler seems to you? Im asking cause some other manufacturers have trouble creating fast UEFI (eg. it lags, reacts very slowly and you can even end with reseting BIOS cause your PC crash faster than UEFI loads).

So its pretty important for OC.

[PiLsY]

The bios is very quick - no lag whatsoever in use and fast post. Post to OS in around 30 seconds on a really old sata hdd, 9 seconds on a G.Skill Falcon SSD.

[blindfitter]

what the problem with unlocked cpu do you mean
is this with QF side or the normal foxconn side

Carl,

I think what dave was on about, is that they default to 60multi if using unlocked multiplier chip! 775 did

[Dualist]

Carl,

I think what dave was on about, is that they default to 60multi if using unlocked multiplier chip! 775 did

That was the MSI Dave.

Carl, the Bloodrage had issues with the unlocked multi and going to a max of 25 multi when you disabled EIST (or C1E I forget), you could run unlocked if you had this on but it dropped (as it should) but not great when you're under ln2.
I played with turbo and everything else, best I could get was an unlocked multi but one less than I set in the bios, then it jumped up again to what was set. Can't remember all the settings because I got rid of the board about a year ago but was the only reason to sell it, loved it otherwise.
It also did the same with 975's, why was this the only board to do this.?
Sorry for O/T but was just gonna pm you but got sidetracked.
Dave.

[Lord Trident]

think that was old bios of BR and if it not reported it can not be fixed



i know some of the normal foxconn have problem with unlocked cpu
but have reported but nothing so dont both anymore

[jstutman]

Does foxconn have a release date for this board?

[Mescalamba]

Um.. its already released I think? Though it will take maybe a bit time until you will be able to buy it. Im slightly tempted, but I guess it wont rival my setup.

[jstutman]

Um.. its already released I think? Though it will take maybe a bit time until you will be able to buy it. Im slightly tempted, but I guess it wont rival my setup.

This /\

The amount of P67 board in stock at most vendors are dwendling (sp?)
Is it cause of the next x68 chipset coming out in May?

[Mescalamba]

This /\

The amount of P67 board in stock at most vendors are dwendling (sp?)
Is it cause of the next x68 chipset coming out in May?

Kinda no idea what you mean by "/\".

P67 boards in my country are well stocked, maybe too well, they dont sell that much of them.

Dunno about X68 .. I saw some news about X79, which should replace X58. I admit Im bit confused by Intel. Though there will be some new chipset soon.

[Hondacity]

great review pilsy

goods
+read pts
+bios looks very deep
+temp displays
+hypersssss

retail price?

[l0ud_sil3nc3]

I am glad to hear it clocks hypers

[Hondacity]

thanks for seeing that ...

[-Radiant-]

i would love to buy this board but... IDE?! and the overclock buttons tweak BCLK not the multi ?? add 2 sata instead of the 300 year old ide and fix the oc buttons failure and i'll buy one

http://www.vortez.net/articles_pages..._review,1.html

[PiLsY]

How often do you need to adjust multi on the fly? Far better to be able to tweak bclk given youre working at high multipliers anyway. Have requested that the buttons be fine step of 0.1mhz instead. Realistically speaking youll be doing more bclk tweaking than multi tweaking. Multi is pretty much set and forget.

Adjusting multi on the fly would play havoc with dynamic vid / turbo etc. If it was easy enough and worthwhile enough both could have been incorporated as im told the circuit is programmable.

Thread is not dead - have been doing some pretty boring software bug hunting and am now awaiting a new bios .

[Lord Trident]

well you can not set muliplyer on the fly
as you have to reset bios each time

so this is the reason for it on the bclk

and on bclk you could get higher clock speed when the multipler does not work
say you can not do 50x 100 which is 5ghz so you set to 49 and change the bclk to 103 you can have 5047 so you will be happy

at the same time say you just need a few mhz to have you first new record on overclocking they again changing bck on thhe fly will help

[PiLsY]

New bios arrived .

Looking very tidy so far. This is just bug fixes + cpu pll override. More mem timings to come.

Dynamic voltage setting is something incredible. If you've ever been through the pain of offset overclocking this will be a total godsend. Dynamic sets default vid and adds your manual offset to it ignoring all cpu turbo vid tables, so in other words the vid scale reacts exactly as if it was at stock turbo conditions and simply adds your desired voltage to it under load.

The QF PWM is working very well too. No LLC to adjust on this board and you dont need it. High cpu speed stability is streets ahead of the P8P67 Pro.

Quick example - 4.8ghz on the P8P67 Pro requires 1.38v manual and extreme llc and switching giving 1.396v under load. The same 4.8ghz on the Rattler in Dynamic mode requires an offset of +90mv for 1.342v under load and 6c lower cpu temps. Its still scaling as well where the Asus board was requiring ever increasing amounts of vcore for prime blend stability. 4.9ghz on the asus required 1.43v under load - 4.9ghz is currently priming at +120mv dynamic offset on the Rattler which gives 1.366v under load. Kerching!

[D0m|nat0R]

http://www.pcaxe.com/hardver/maticne...oxconn-rattler

[shadow_boxer]

Nice looking board. Thanks for posting.

[PiLsY]

New bios is leaps and bounds better than the old.

Prime @ 50x100 +200mv dynamic (gives 1.43v single core loads 1.377v 4 core load) with only 1.106v uncore. Going just over 4 hours now. Blown the asus out the water on cpu clocking (previous best 4.94ghz @ 1.48v 1 hour prime before dropping). Once I figured out that the acpi standby/hibernate and s4/5 power on settings had to be disabled with Internal PLL Overvolt enabled stability has been perfect (c-states still enabled tho).

When using manual voltage or standard + offset it performs just like the asus Pro in terms of cpu clocking so its definitely down to Foxconns implementation of their own hardware controlled Dynamic voltage option. Impressive stuff.