Welcome Guest ( Log In | Register )

Bump Topic Topic Closed RSS Feed
7 Pages « < 2 3 4 5 6 > » Bottom

Outline · [ Standard ] · Linear+

 AMD® Socket AM2&AM3 Overclocking V24, Thuban or Bulldozer!? Its 6-Season

views
     
dafreak
post Mar 6 2011, 09:51 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(saturn85 @ Mar 6 2011, 07:50 PM)
oh, i thought AM3 socket have 938 pins only instead of 941 pins?
so that a AM2+ CPU with 940 pins cannot install on it. unsure.gif
*
the holes for the pins cannot match AM2+ procs
dafreak
post Mar 9 2011, 07:33 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE
AMD Bulldozer, Llano Launch Dates Leaked

Various sources point to June and July release dates for Bulldozer and Llano.

ZoomBased on information gathered by "industrial sources," various reports are indicating that AMD plans to launch its upcoming high-end desktop Bulldozer CPUs sometime during the week of June 20, 2011.

According to the reports, Bulldozer will arrive in quad-core, six-core and eight-core variants using the FX brand name. Two will be AMD FX4000-series chips, two will be AMD FX6000-series chips and four will be AMD FX8000-series chips. Four processors are slated to arrive in Q2 2011 while the other four are planned for a Q4 2011 release.

After the release of the initial Bulldozer processors in June, AMD is expected to launch its Llano "A-Series" APUs. These will arrive on the market in two-core, three-core and quad-core variants during the week of July 4. A total of eleven Llamo APUs will be released throughout the calendar year, the reports claim.

An additional leaked slide maps out the next two years, revealing that all three Llano APUs will supposedly support 1,600 MHz DDR3 memory at 45W and feature the "BeaverCreek" and "WinterPark" integrated DirectX 11 GPUs. The Llano APUs will target the mainstream and performance markets during 2011 whereas the Zacate and Ontario will focus on the essential and sub-12-inch markets respectively.

Looking ahead into 2012, AMD's roadmap shows the Trinity, Richland, Wichita and Krishna APUs. The first three will target the essential, mainstream and performance markets while the Krishna APU will target the sub-12-inch devices. All four feature Fusion graphics and support DDR3 memory.


nom nom sos
dafreak
post Mar 10 2011, 08:51 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(lengchai86 @ Mar 10 2011, 08:33 PM)
saturn85 : isit 1.45 is too high? because what i understand is this proc C3 dont need a high voltage to get stable.. do correct me if i'm wrong

kuntawakaw : what do you mean 'dram controller option'? i set to ganged mode. ganged mean dual channel right? what is 'llc'?
*
llc = load line calibration

ganged/unganged mode also dual channel ler
dafreak
post Mar 12 2011, 11:40 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(AMDAthlon @ Mar 12 2011, 10:27 PM)
Prepare for the Bulldozer.The boxes can poison u to buy it whistling.gif
» Click to show Spoiler - click again to hide... «


http://www.amdforum.se/artikel/exklusivt-a...-och-fx-4-1025/
*
FUUUU! BD chips not out yet box sudah ada laugh.gif

not the usual black anymore drool.gif
dafreak
post Mar 14 2011, 12:24 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE
Four AMD Bulldozer Chips Incoming: Details Revealed.

AMD FX "Bulldozer" Approaching, Set to Consume 125W, 95W
[03/13/2011 11:36 AM]
by Anton Shilov

The initial family of AMD's FX-series microprocessors based on Bulldozer micro-architecture will include four models with eight, six or four cores and 95W or 125W thermal design power (TDP).

The first breed of AMD FX8000, FX6000 and FX4000 currently known under Zambezi code-name will completely support all the advantages that the Bulldozer micro-architecture is supposed to bring, including new Flex FP floating point processing unit. The new chips in maximum eight-core configurations are projected - by AMD's internal documents - to offer roughly 50% performance improvement over Phenom II-series microprocessors in multimedia applications.....


delicious sos drool.gif

QUOTE
AMD Aims to Fight Core i7 “Sandy Bridge” with Bulldozer.

AMD Wants to Increase Pricing of Microprocessors
[03/13/2011 04:37 PM]
by Anton Shilov

Advanced Micro Devices at present is not able to compete in terms of pure performance against Intel Corp.’s Core i7-series “Sandy Bridge” microprocessors, according to benchmark results, but with its next-generation Bulldozer-based chips the company fully aims to fight.

With its FX-series AMD Zambezi desktop processors the Sunnyvale, California-based company plans to fully compete against Intel Core i-family of high-performance central processing units. A document from AMD seen by X-bit labs claims that AMD FX-series central processing unit (CPUs) powered by the Bulldozer micro-architecture will be fully able to rival Core i7 2600-series chips.....


MOAR!!

This post has been edited by dafreak: Mar 14 2011, 12:26 PM
dafreak
post Mar 27 2011, 10:05 AM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


lol wtf? must be fake, no way 4v the proc can survive doh.gif

4v for 3.8ghz rofl, mine dont even need that much
dafreak
post Mar 30 2011, 09:36 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(eNOOBz @ Mar 30 2011, 09:32 PM)
I'm using Corsair XMS3 1600MHz RAM but in the CPUZ shows the max freq is only 663MHz. Why owh?
*
663 x 2 = 1326
dafreak
post Apr 4 2011, 09:36 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(cllee86 @ Apr 4 2011, 07:53 PM)
Hi guys. Latest update on my OC adventure.

Prime 3.5Ghz @ 1.425v = failed - PC restarted in less than 3 seconds sweat.gif 

Prime 3.2Ghz(Stock) @ 1.375v(Under volt by 0.025v = failed - PC also restarted in less than 3 seconds sweat.gif

At last, Prime 3.4Ghz with stock voltage @ 1.4v - Successful for 45mins without any issues. Maximum temperature reach 62c sweat.gif 

Aiii...in the end the most stable clock speed is the one I've been using all along doh.gif   Feels kinda disappointed sweat.gif 

I can try 3.6Ghz @ 1.45v but the temperature will climb very fast and will exceed 65c in no time. My biggest concern is heat now.
Gotta find some way to reduce the temperature of my CPU and improve the air flow in my casing.

Maybe need to radically mod my casing tongue.gif
*
mine only need <1.4v for 3.6Ghz stable sweat.gif

i suppose you need a better cooler

This post has been edited by dafreak: Apr 4 2011, 09:36 PM
dafreak
post Apr 4 2011, 10:04 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(cloudwan @ Apr 4 2011, 09:52 PM)
It's either cooler or he badly need a better PSU liao, Enlight 400W is ok for me for light~medium load coz i used Enlight 425W b4 but that was during X2 3600+ yoo.. how many years was that.. sweat.gif
With u oc'ing ur 955W which is a 125W TDP cpu + quite power hogging HD4870 i think its a bit too much to ask of that PSU liao.. hmm.gif
*
didnt notice that...could be this also..

try borrowing a better and higher wattage psu and try hmm.gif
dafreak
post Apr 4 2011, 10:26 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


cap ayam psus are known to be able to deliver only half their rated wattage..and inefficient anyway...

get at least an 80+ rated psu or better 80+ gold

if you want to isolate the problem, try running barebones


Added on April 4, 2011, 10:37 pmyour psu is underpowered...

if you dont believe you can input your components here

with your components, i get recommended psu wattage 376W

means you need at least a 500W psu, or a 450W@80+gold

This post has been edited by dafreak: Apr 4 2011, 10:37 PM
dafreak
post Apr 4 2011, 10:43 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(cllee86 @ Apr 4 2011, 10:41 PM)
Can't really call Enlight cap ayam. Have dissected the PSU before and it was surprisingly good. Large transformer, surprisingly quality large capacitors and some very good transistors too.

Oh, did I mention it was as heavy as some of the 550w PSU's biggrin.gif
*
heavier than a 550W psu doesnt mean it can deliver as much as a 550W, larger stuff doesnt mean more efficient/better

btw, read 2nd post smile.gif
dafreak
post Apr 4 2011, 11:18 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(cllee86 @ Apr 4 2011, 10:51 PM)
Nah..those PSU calculator are not very accurate. Just another tool to make money.

You see. There is a strong correlation between weight and a quality of PSU. Normally, a more powerful PSU requires a larger transformer, which in turns translated to heavier PSU.

And there is no way these tiny copper wire you found in those high end PSU can carry 60amps in them. Unless they're made out of pure gold or the found a way to turn it super conductor at high temperature biggrin.gif 

You don't really need that much power in a small CPU you see. Its only 1.4v for God sake sweat.gif

And if my PSU is really under power, it would have shut down a long long time ago biggrin.gif
*
psu delivers power to the whole system, not just the cpu

1.4v may seem small but the amount of power the cpu needs is quite large especially when you are testing for stability (e.g. prime95)

better isolate by running barebones smile.gif
dafreak
post Apr 4 2011, 11:29 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(cllee86 @ Apr 4 2011, 11:20 PM)
Well. I appreciate your kind intentions but if you did as much research as I did into PSU then you'll know that the ampere rating on those 12v rails could be best describe as the work of fiction.

There is 1 time I was obsessed by PSU and all these 12v rail rating. I'm gonna take my time to explain why 60amp on those PSU 12v rails are crazy.

Our PC runs on Direct Current. Which means that the positive and negative charge must go together in the same copper wire. This creates a more stable flow of electricity but generates a lot of resistant. Which means, in order to carry more current (known as ampere) it must have thicker copper wire.

Now if you need a copper wire to carry 60amps of 12v DC, you'll need at least 1" thick copper wire sweat.gif 

Have you ever see a 1 inch thick cable in any PC PSU? Neither did I. But I have seen a 12v DC 40amp PSU, but not for PC. Its for Car starter and you'll need a trolley just to move it tongue.gif 

Now, about wattage calculation. Watt is a very general term. It can be used to represent any form of power. Even the power required for a person to climb stairs can be measured in watt. Now for electricity, watt is easier to calculate. To get the wattage, you just multiply the volt and the ampere.

V * Amp = Watt, for example.

10v * 10amp = 100watt. Easy. But if you take the formula to calculate CPU then it all doesn't make sense. CPU is only 1.4v, so in order for it to consume 125watt it must take at least 89amps shocking.gif  shakehead.gif 

So for a tiny CPU to consume that much power is sweat.gif 

I still need some time to find the root cause for my OC setup. But I'm confident that the PSU is not the culprit.

Btw, I'm more interested in the settings. Like how will the NB frequency affect OC and other stuff like that..
*
bro, that's just the cpu core only doh.gif there are other parts on the cpu also lar (e.g, IMC, NB, others that we cant change volts on)

why would psu makers put a 'fictitious' rating on their psu anyway??? rofl doh.gif
dafreak
post Apr 6 2011, 08:49 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


longer to complete as to compared to what settings?
dafreak
post Apr 6 2011, 09:53 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


try using 1T instead

p.s. post the pics so we can compare easily smile.gif
dafreak
post Apr 7 2011, 10:04 AM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(shajack @ Apr 7 2011, 09:59 AM)
like cllee said,which is better...higher freq or tight timing?
*
obviously is both... smile.gif

superpi/hyperpi dont just depend on ram, still depend on other factors like cpu speed, nb speed

your rams may be overclocked, but may be bottlenecked by other factors above unless they are OC-ed as well
dafreak
post Apr 7 2011, 10:45 AM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(cloudwan @ Apr 7 2011, 10:08 AM)
Not NB speed yo but cpu-nb speed smile.gif
*
cpuz show as NB speed mar tongue.gif laugh.gif

bios is as cloudwan says, cpu-nb speed
dafreak
post Apr 7 2011, 06:29 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


QUOTE(kuntawakaw @ Apr 7 2011, 06:07 PM)
u will end up disappointed. why u take high freq (i assume 2k and above) but most of rams rated on intel. and u cant get it run at 2k u will curse ur ram cuz cant even run at rated. dont even mention tightening the timings.
but if u buy a 1600mhz ram, and it runs at 1800mhz then u feel good already..cuz i will run more than rated..kekeke.thats why last time i like value rams.. 1333>1900mhz makes me feel good  laugh.gif  laugh.gif
*
psychology lor...hehe...want to get >2k mhz ram and run @ rated get inhell lo whistling.gif

@cloudwan: we garu ram ada la laugh.gif
dafreak
post Apr 7 2011, 09:11 PM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


IMO, prime95 is the best to test mem

memtest may pass, but can fail prime95
dafreak
post Apr 9 2011, 01:53 AM

Regular
******
Senior Member
1,064 posts

Joined: Jul 2008
From: Star Stuff


if i were you, i wouldnt consider 17 minutes of prime = stable

maybe stable enough to bench but not stable for daily usage


Added on April 9, 2011, 1:56 amtry loosening the timings abit to 6-6-6

This post has been edited by dafreak: Apr 9 2011, 01:56 AM

7 Pages « < 2 3 4 5 6 > » Top
Topic ClosedOptions
 

Change to:
| Lo-Fi Version
0.0572sec    0.77    7 queries    GZIP Disabled
Time is now: 19th December 2025 - 03:52 PM