Nice job, BOlts :cool:

Thanks! This should help anyone wanting the latest RIVE drivers!

Hi BOLTS4BREKFAST,

First of all thanks for your work gathers the most recent drivers for RIVE.
I wanted to ask for info on your way to install the Intel Cipset because I do not understand why. Why advice to use the procedure ...instead of the simple setup inside the new folder? Is there a particular reason?

"2)Go into CHIPSET folder
3) Open "Intel Chipset (9.4.0.1027) NEWEST" folder
4) Hold down SHIFT + RIGHT MOUSE CLICK in that folder and select "OPEN COMMAND WINDOW HERE". That will open up a command prompt in that folder
5) In the command prompt type "SETUP.EXE -OVERALL" and hit ENTER..."

Finally a question I meant to get you at the first opportunity, since I saw your name on the scoreboard-Legue of RealBench, close to mine (731 for you, 730 for me): I have always surprised at the so small difference that exists between my hardware that significantly below your own, apart the use of the same type of memories Corsair.
It seems impossible that a 3960X CPU @ 4700, liquid cooled, a tree-SLI EVGA GTX 780, and a Raid0 of very fast Samsung 840, are able to obtain "ONLY one point more" than my poor 3930K @ 4700 air cooled, a single old HD6970 video card, and a Raid 0 of miserable Vertex4, .... there is certainly something wrong on tests of RealBench, or about your setup?
What do you think? Do you have any other explanation?



greetings
sandro c.

RealBench is not really affected by GPU or SSD differences. It mainly wants faster CPU clocks and faster RAM frequencies and/or tighter timings. If you are both running similar frequencies for CPU and RAM without really tweaking the OS or timings the scores will be similar.

BOLTS! very nice resource to post up +1:)

Hi Arne,

Just this may be the real explanation, if the test not does not act as a variable speed of writing and reading Mem-hdd or the GPU's processor ... so if our CPU clock are both @ 4700, certainly earned my clock memory to 2180, while the other was probably a stock-speed of 2133.
If so, then 50 Mhz more of my memory speed have produced ... one point less?
Hello Arne
sandro c.

Yes quite possibly, especially if timings are different and making you pretty equal. And maybe one of you was running slightly more in the background etc.

That is why it is called Realbench...because in the real world any modern SSD will perform pretty much like any other. Performance gains are really CPU and RAM frequency increases (and tight RAM timings). Second and third timings count...

The next version of RealBench will have GPU test etc.

Hi Arne,

Very clear Arne, but What I did not understand what you mean by "(Second and third timings count ...)" if for second mean value of the second timing or ... SECONDARY timings.
Our memories are similar but not identical: you use platinul 2133, I use D.GT 2133.
I do not remember to whom and when, but in a tread you said that you use your DDR with these times:

for your 2133 :9-11-10-30-2t / 7-183-18-9-27-9-8 / 0-2-3-5-5-3-1-1-5-0
for your 2400 9 - 9 - 9-24-1t / 5-107-10-5-20-5-7 / 0-2-3-3-2-2-1-1-2-0

at 1.5v

for my 2200 :9-11-10-27-1t / 4 - 97-9 -4-20-4-8 / 0-0-0-0-3-3-1-1-3-0
for my 2400: I could never run @ 2400 ...!

at 1.55v
I have a Mem.bandwidth of 55.80 by Sisoft sandra and dram score of 46766 by MemTweak
You see these timing something wrong?

Thanks Arne.
sandro c.

Sorry sandro c yes I meant secondary...tertiary...timings.

I got 2400+ to run 9-11-11-24-1T nearer/4-96-9-4-16-4-8/0-0-0-0-3-3-1-1-3-0 in the end....at 1.71v

Nothing wrong with those timings at all...at least not to my untrained eye...look very good...very similar to my own...so kits not so different 🙂