Overclock.net › Forums › Components › Memory › Memory: Why Pay More?
New Posts  All Forums:Forum Nav:

Memory: Why Pay More? - Page 6

post #51 of 66
Thread Starter 
Quote:
Originally Posted by SpcCdr View Post
Why pay more?
(A rational response- not "tossing out troll-bait)
The better quality memory is worth the money because it is simply better quality memory with better quality manufacturing: (for the most part- some bad sticks still do get through quality control- regardless of brand)- and most higher quality memory is also offered in higher rated speed kits:
Luck at getting a decent OC -vs- buying a kit that is guaranteed to do that speed.

The quality of the Silicon is better; the screening is better during etching; the resulting IC's are better binned, and the final modules are generally hand selected and tested to a higher frequency; ergo, The manufacturer will guarantee the kit to run at the higher "stock" frequency. (Be careful not to confuse "stock" with Factory OverClocked!)

To clarify a little further:
Th OP (Paraphrased)

So, it's Very lucky to get a "stock" (read as "cheap") 800 MHz kit to run at 1066, as opposed to a kit that has been factory overclocked and guaranteed by the manufacturer to run at 1066 MHz (Still not "stock")!

While we're at it and discussing this (and the OP (and others) did bring up G.Skills ), let's sidetrack and take a look at the three popular G.Skill kit's that everybody seems to get confused over:
G.SKILL 4GB (2 x 2GB) DDR2 800 (PC2 6400) F2-6400CL5D-4GBPQ ($76.99)
G.SKILL 4GB (2 x 2GB) DDR2 1000 (PC2 8000) F2-8000CL5D-4GBPQ ($89.99)
G.SKILL 4GB (2 x 2GB) DDR2 800 (PC2 6400) F2-6400CL4D-4GBPK ($104.99)

G.Skill's naming convention Is (P)erformance (K)ing trumps (P)erformance (Q)ueen (Quality of Silicon, PCB, etc.)
(and obviously Cas Level 4 is better than CL 5. )

So the PKs are better binned than the equivalent PQs
DDR2 800 PK > DDR2 800 PQ Right?! Yep, Sorted!

What about the 1000MHz PQ's vs the 800MHz PK's??
We know PK tops PQ, but, 1000MHz beats 800MHz Right?!

and yet 800MHz CL4 loosened to CL5 would probably get you to 1000MHz !
So that would be 800MHz PK's OC'd to 1000MHz (Still PK) -vs- 1000MHz ("stock"/"Factory OC'd) PQ's

We Have ourselves A conundrum!

*Official* Response from G.Skill:

So, You could OC the 800PK's yourself for Maybe a set of 1000MHz PK's OR purchase the 1000PQ's and be guaranteed 1000MHz (albeit on PQ's)
*Imagined last October* "If they only made 1000MHz CL4 PK's"*
I asked them Again; "What about the difference between 800PK and 1000PQ?" (Emphasis on Quality aspect not speed)
"They are Same PCB for these two set" (sic) ..."However, the 1000's are tested, binned, and hand selected....."

*Imagined last October* "If they only made 1000MHz CL4 PK's"*
Now they Do!(Hint1066MHz PK's)

Phew, now that's cleared up Again!
Back to the OP's follow up clarification:


There is no such thing as "stock 1066" memory! it is ALL Factory Overclocked.
So, from the "Truth in Advertising" perspective:
-800MHz "Factory OC'd to 1066" is the "real deal" and any company that sells their memory as 1066 "stock" is "telling you porkies".

So, we buy the 800MHz set and pray to the almighty RAMa that we can get them to 1000MHz!
Maybe we can, Maybe Not!, Maybe they'll hit 1000 and even go a little further! )
or, we buy the 1000's and know that they will do 1000MHz "stock"! -as the factory has previously tested this "Stock 800MHz" kit to be of a higher quality and capable of running at a stable 1000MHz "Factory Overclock"
-and I'm sure they will be just as likely to go that bit further over their tested parameters, as the original "lower quality" 800's were to hit 1000MHz.

As far as your Kit, you might have found a rare gem but, I'd be wary, the quality of components (Not just the Silicon; also, the FBGA solder, the landings, the PCB etc.) Might not be up to scratch.

Please, have a read through these: (I would recommend bookmarking and pacing yourselves- they're quite "Ooh- no more, no more, my brain hurts") (Mine did )
The one most pertinent to this thread is Part 2, but, they should all be "required reading"!
Sectets of PC Memory Part 1 -The basics
Secrets of PC Memory Part 2 -DDR Form Factors
Secrets Of PC Memory Part 3 - Memory Generational Differences
There is also a Secrets of PC Memory Part 4 -The Next Gen. (DDR3)


Cheers to ALL!

[edit] seems like I missed quite the little argy-bargy while I was culling research links for you guys!
That's a bunch of great info, but doesn't that basically confirm my original theory? From what you've said, Kingmax has tested and stated the sticks to work at 1066 (regardless of them being cheap or not) vs. a much more expensive, much higher quality, G.Skill 800 kit that only might get to 1066. If I'm reading it correctly, if the manufacturer says it'll work at 1066, that's better than a kit that is only guaranteed do 800 regardless of the quality of silicon/manufacturing process. I didn't think it would spark this much of a debate. Perhaps my question was a bit convoluted.

Either way, thanks to all those who replied. For me, the proof is in my benchmarks and my wallet. This cheap memory is running cooler, with less voltage (1.8V vs. 2.15V), and at the EXACT same speeds as my G.Skill kit that cost more than 3 times as much. The funny part is, the only reason I picked up these cheap sticks was because my G.Skill kit made from quality silicon and all those other bells and whistles died on me.
RAID0R
(14 items)
 
  
CPUMotherboardGraphicsRAM
i5 750 4.0GHz MSI P55-GD80 GTX 470 | 8800GT PhysX 2x2GB G.Skill Ripjaws 
Hard DriveOptical DriveCoolingOS
60GB Agility 2|1TB RAID0|1.5TB Pioneer DVR-217D XSPC Raystorm | XSPC RX240 Windows 7 Professional x64 
MonitorKeyboardPowerCase
27" Dell 2709W | 17" Samsung Logitech G15 Corsair HX850 Corsair 650D 
Mouse
Microsoft IntelliMouse 
  hide details  
Reply
RAID0R
(14 items)
 
  
CPUMotherboardGraphicsRAM
i5 750 4.0GHz MSI P55-GD80 GTX 470 | 8800GT PhysX 2x2GB G.Skill Ripjaws 
Hard DriveOptical DriveCoolingOS
60GB Agility 2|1TB RAID0|1.5TB Pioneer DVR-217D XSPC Raystorm | XSPC RX240 Windows 7 Professional x64 
MonitorKeyboardPowerCase
27" Dell 2709W | 17" Samsung Logitech G15 Corsair HX850 Corsair 650D 
Mouse
Microsoft IntelliMouse 
  hide details  
Reply
post #52 of 66
c-bro now can you do us a favor? post a link of where you bought them at and if you can see the chips tell us what kind they are or post a pic of them, and last thing is give us a review of how high you can overclock them
compy jr
(13 items)
 
  
CPUMotherboardGraphicsHard Drive
xeon 3350 at 3.6ghz Asus P5k-dlx HIS 7970 @ 1150/1900 Lots of them.  
OSMonitorKeyboardPower
vista ultimate x64/xp sp3 compaq 21" junk dell keyboard corsair tx850m 
CaseMouseMouse Pad
upgraded metal box You don't need a mouse Desk 
  hide details  
Reply
compy jr
(13 items)
 
  
CPUMotherboardGraphicsHard Drive
xeon 3350 at 3.6ghz Asus P5k-dlx HIS 7970 @ 1150/1900 Lots of them.  
OSMonitorKeyboardPower
vista ultimate x64/xp sp3 compaq 21" junk dell keyboard corsair tx850m 
CaseMouseMouse Pad
upgraded metal box You don't need a mouse Desk 
  hide details  
Reply
post #53 of 66
Quote:
Originally Posted by rx7speed View Post
SpcCdr if a company is selling ram at 1066mhz then by all means the ram should support and run at that speed. to do so otherwise would be along the lines of advertising their product falsely. So with this Kingmax memory being sold as 1066mhz ram he isn't lucky to have it run at 1066mhz. now granted he might be lucky to run it at 1070mhz I won't deny that but it still should run at it's rated speed.

let me ask you this. If I bought 667mhz ram and it peaked out at 750mhz would the ram be defective? would I be able to get any recourse of this? doubtfull as it ran at it's rated speed plus some. But if I bought ram that was rated at 1066mhz and it only 850mhz then would the ram be deemed defective? I would say so. I'm also sure I would be able to get recourse over something like this if all their product came in at the same situation.

that is what kingmax seems to be doing here is selling 1066mhz ram so why is it luck that it is able to run at it's rated speed?
Yep Rx You are 100% correct, If they're rated at 1066, then by gox they should do 1066, and if they don't RMA 'em. It is extremely rare to find "cheap" RAM that is rated at 1066 due to the higher quality of production needed to achieve that rating. That's where the luck comes in!
(I'm thinking Very lucky - to get a set of 800's to1066.- prolly only cuz they were "high quality" HZ's,
I'd think it would be extremely lucky to get a set of 800PK's or PQ's anywhere near that high)
The other part of the original explanation was that his "higher quality" G.Skill 800's were OC'd (by him)- to a much higher speed than what the manufacturer rated and guaranteed them to. (The other part regarding his luck- some kits just simply will not go much above what they are rated) (Which is exactly what we both said)
Quote:
Originally Posted by C-bro View Post
That's a bunch of great info, but doesn't that basically confirm my original theory? From what you've said, Kingmax has tested and stated the sticks to work at 1066 (regardless of them being cheap or not) vs. a much more expensive, much higher quality, G.Skill 800 kit that only might get to 1066. If I'm reading it correctly, if the manufacturer says it'll work at 1066, that's better than a kit that is only guaranteed do 800 regardless of the quality of silicon/manufacturing process. I didn't think it would spark this much of a debate. Perhaps my question was a bit convoluted.

Either way, thanks to all those who replied. For me, the proof is in my benchmarks and my wallet. This cheap memory is running cooler, with less voltage (1.8V vs. 2.15V), and at the EXACT same speeds as my G.Skill kit that cost more than 3 times as much. The funny part is, the only reason I picked up these cheap sticks was because my G.Skill kit made from quality silicon and all those other bells and whistles died on me.
There's always the chance of getting a funky chunk of SiO, and it may also have been something to do with running your 800MHz set at 1066 . GL with your RMA'd set.

Bingo, If they say 1066 then "Battle of Hastings" it had better be 1066.. (See above) but we were originally comparing the OC ability (Luck vs Guaranteed speed) of an 800MHz kit, OC'd to 1066, and a kit sold as 1066 and guaranteed to run at 1066 by the manufacturer.
Do you think if you bought a 800MHz set of these KingMax (cheaper still than the 1066 set), You'd be able to get them to 1066?
I'd say "Lucky" if you did!
But, Prolly not.
That was the whole point of the question/answer.
Royally MIIFed
(13 items)
 
  
CPUMotherboardGraphicsRAM
Q9650 @4.43GHz (492FSB) MIIF 465GTX (GE) & GTX 260+ G.Skill 1100PI (4x2)@1181 
Hard DriveOptical DriveOSMonitor
80G SSD; 150GB VR; F3 1TB; 2x500GB F3s RAID0 LG BluRay/Sammy 22x Multi W7U64 HP lp2475w (H-IPS) 
KeyboardPowerCaseMouse
G11 CM SPG 1200w CoolModded CM690 G7 
  hide details  
Reply
Royally MIIFed
(13 items)
 
  
CPUMotherboardGraphicsRAM
Q9650 @4.43GHz (492FSB) MIIF 465GTX (GE) & GTX 260+ G.Skill 1100PI (4x2)@1181 
Hard DriveOptical DriveOSMonitor
80G SSD; 150GB VR; F3 1TB; 2x500GB F3s RAID0 LG BluRay/Sammy 22x Multi W7U64 HP lp2475w (H-IPS) 
KeyboardPowerCaseMouse
G11 CM SPG 1200w CoolModded CM690 G7 
  hide details  
Reply
post #54 of 66
Thread Starter 
Quote:
Originally Posted by rx7speed View Post
c-bro now can you do us a favor? post a link of where you bought them at and if you can see the chips tell us what kind they are or post a pic of them, and last thing is give us a review of how high you can overclock them
http://www.canadacomputers.com/index...id=RAM.346.754

From the manufacturer, it's a matched pair of these:

http://www.kingmax.com/products/dp/ddr2_1g_1066.htm

The real chips actually say Kingmax on them.

Everything about these sticks screams ghetto, but I could care less. Just don't go buying all of them up. At that price I may just go to 4GB. I have a copy of Vista Business X64 collecting dust anyway.
RAID0R
(14 items)
 
  
CPUMotherboardGraphicsRAM
i5 750 4.0GHz MSI P55-GD80 GTX 470 | 8800GT PhysX 2x2GB G.Skill Ripjaws 
Hard DriveOptical DriveCoolingOS
60GB Agility 2|1TB RAID0|1.5TB Pioneer DVR-217D XSPC Raystorm | XSPC RX240 Windows 7 Professional x64 
MonitorKeyboardPowerCase
27" Dell 2709W | 17" Samsung Logitech G15 Corsair HX850 Corsair 650D 
Mouse
Microsoft IntelliMouse 
  hide details  
Reply
RAID0R
(14 items)
 
  
CPUMotherboardGraphicsRAM
i5 750 4.0GHz MSI P55-GD80 GTX 470 | 8800GT PhysX 2x2GB G.Skill Ripjaws 
Hard DriveOptical DriveCoolingOS
60GB Agility 2|1TB RAID0|1.5TB Pioneer DVR-217D XSPC Raystorm | XSPC RX240 Windows 7 Professional x64 
MonitorKeyboardPowerCase
27" Dell 2709W | 17" Samsung Logitech G15 Corsair HX850 Corsair 650D 
Mouse
Microsoft IntelliMouse 
  hide details  
Reply
post #55 of 66
In this case (the OPs original post) there is near zero reason to get the the G.Skill at that price.

Kingmax is not a bad brand. And I would rather have a set of Kingmax 1066 than most G.Skill 800, especially considering the price difference.

Also, kingmax does not make DRAM ICs. They buy them and relable them, just as many other companies do.

Quote:
Originally Posted by The_Rocker View Post
I say it again...

Because it looks and sounds cooler.

I'd rather have 4GB of Crucial Ballistix DDR2 Memory with gold and silver heatspreaders than 4GB of generic exposed chip crap.


It is also made from better quality / ocing chips.
Dumbest. Reason. Ever.

If the superfical appearance is that important to you, you can spend an extra 10-20 dollars and make the ram look like whatever you want it to.

Quote:
Originally Posted by Chozart View Post
I don't like heatspreaders. They indeed do isolate... I think there is a Tomshardware article about it. The OCZ design seems an exception though.
Quote:
Originally Posted by Perry View Post
This is why I love my OCZ sticks. Yes, they have a heat spreader but it's an open design so it adds additional heat conductive surface area but doesn't allow warm air to be trapped in between the PCB and the metal of the heat spreader itself.

Observe...



Yes, you can see the IC's but what you're seeing is thermal tape to keep the heat spreaders attached. The area in between the IC's is open and able to come in contact with air directly.
Soild metal heatspreaders, as little as they typically do, are better than the OCZ design. If you are going to put a heatsink on something, you are doing so because air itself does not have the thermal conductivity to remove enough heat from as small an area as you want to cool. You do not want holes in your heatsink/heatspreader exposing the ICs/die to air. You want the most dense material with the highest thermal conductivity possible in contact with the ICs; then you want to give that material as much surface area as possible.

The OCZ heatspreaders are possibly the worst design of all. With the OCZ "heat spreaders" you have the ICs, thermal tape, and an open honeycomb.

-The total surface area of the heatspreaders is not significantly larger than that of the ICs (a problem with most heatspreaders), so even with zero thermal resistance and infinitly high thermal conductivity (both impossibilities in this case), you would, at best, have the exact same temps as with no heatspreaders at all.
-The ICs receive no cooling on most of their surface area, as the heatspreaders are full of holes.
- Even where there are holes, there is thermal tape, further increasing thermal resistance.

Anyway, a decent set of heatspreaders will help more by evening out temperature hotspots than by lowering overall average temperatures. I am quite sure the average temperature of my memory is not much different with heatspreaders than without (given the rather thick interface), but without them, a handful of chips get hot enough to burn skin, while others are barely warm to the touch. Having half the chips om my dimms last 10 years while allowing the other half to fail in 3 months is no good. I would much rather have them all die in 2-3 years, as a dimm is only as good as it's weakest IC.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
post #56 of 66
always wondered too,
Bought 2x1Gb of DDR2 800Mhz ram, 4-4-4-12 stock. Hynix brand i think.

THen came across some OCZ DDR2 800Mhz cheap. (sucker for the Heatspreaders lol), they booted as 5-5-5-15.
Put them to rated 4-4-4-12 and they actually do perform better than the cheap stuff. In everest, the read and write, latency, its all better, not by a huge gap, but enough to notice..

My 2c anyway lol
The DK
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8400 DFI DK P45 T2RS PLUS ASUS EN7900GS OCZ DDR6400 800Mhz 
Hard DriveOptical DriveOSMonitor
1.6Tb worth Lite On DVDRW Vista Ultimate Samsung 226BW 22" 
KeyboardPowerCaseMouse
Generic Coolermaster 650Watt Coolermaster CM690 Generic 
Mouse Pad
Generic 
  hide details  
Reply
The DK
(13 items)
 
  
CPUMotherboardGraphicsRAM
E8400 DFI DK P45 T2RS PLUS ASUS EN7900GS OCZ DDR6400 800Mhz 
Hard DriveOptical DriveOSMonitor
1.6Tb worth Lite On DVDRW Vista Ultimate Samsung 226BW 22" 
KeyboardPowerCaseMouse
Generic Coolermaster 650Watt Coolermaster CM690 Generic 
Mouse Pad
Generic 
  hide details  
Reply
post #57 of 66
Honestly, you pay more for potential The RAM sticks that will come with the Micron D9 chips are the most sought after RAM especially if they have a proven track record for extreme overclocking results. But, actually, there are a few other memories out there that do use Micron D9 chips and are cheap but do Overclock pretty well. A-Data is one of those brands IMO, They have a Vitesta series that comes as 4GB (2GB x 2) DDR2 800 at 4-4-4-12 timings and will even go above 1000mhz. Their Value series with Heatspreaders comes as 4GB (2GB x 2) DDR2 800 @ 5-5-5-18 can actually hit 4-4-4-12 @ 800 and will go close to or above 1000 at 5-5-5-15 which isn't bad for 4GB kit of RAM for under $60.

I used to have a website that listed all the RAM chips on each stick of DDR2 memory made so you could see it before you bought it.

Tomshardware also did a review on like 10 or more Value Series RAM sets that will surprise you.
Kamtastic II
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD Athlon II X4 630 Gigabyte GA-MA790GP-UD4H Visiontek 4850 512MB Reference 8GB A-Data DDR2 800 
Hard DriveOptical DriveOSMonitor
750GB Caviar Black Currently Samsung 20x SH2022 Windows 7 Ultimate x64 2 x Samsung Synchmaster 226BW 
KeyboardPowerCaseMouse
Logitech G11 OCZ 700w GameXStream Thermaltake V9 Black Edition Razer Copperhead 
  hide details  
Reply
Kamtastic II
(13 items)
 
  
CPUMotherboardGraphicsRAM
AMD Athlon II X4 630 Gigabyte GA-MA790GP-UD4H Visiontek 4850 512MB Reference 8GB A-Data DDR2 800 
Hard DriveOptical DriveOSMonitor
750GB Caviar Black Currently Samsung 20x SH2022 Windows 7 Ultimate x64 2 x Samsung Synchmaster 226BW 
KeyboardPowerCaseMouse
Logitech G11 OCZ 700w GameXStream Thermaltake V9 Black Edition Razer Copperhead 
  hide details  
Reply
post #58 of 66
Quote:
Originally Posted by DAN-86 View Post
always wondered too,
Bought 2x1Gb of DDR2 800Mhz ram, 4-4-4-12 stock. Hynix brand i think.

THen came across some OCZ DDR2 800Mhz cheap. (sucker for the Heatspreaders lol), they booted as 5-5-5-15.
Put them to rated 4-4-4-12 and they actually do perform better than the cheap stuff. In everest, the read and write, latency, its all better, not by a huge gap, but enough to notice..

My 2c anyway lol
The subtimings are probably different.

There is no reason for any performance difference if all the timings are actually the same; CAS-tRCD-tRP-tRAS timings are only four of dozens.
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
Primary
(15 items)
 
Secondary
(13 items)
 
In progress
(10 items)
 
CPUMotherboardGraphicsRAM
5820K @ 4.2/3.5GHz core/uncore, 1.175/1.15v Gigabyte X99 SOC Champion (F22n) Gigabyte AORUS GTX 1080 Ti (F3P) @ 2025/1485, 1... 4x4GiB Crucial @ 2667, 12-12-12-28-T1, 1.34v 
Hard DriveHard DriveHard DriveCooling
Plextor M6e 128GB (fw 1.06) M.2 (PCI-E 2.0 2x) 2x Crucial M4 256GB 4x WD Scorpio Black 500GB Noctua NH-D15 
OSMonitorKeyboardPower
Windows 7 Professional x64 SP1 BenQ BL3200PT Filco Majestouch Tenkeyless (MX Brown) Corsair RM1000x 
CaseMouseAudio
Fractal Design Define R4 Logitech G402 Realtek ALC1150 + M-Audio AV40 
CPUMotherboardGraphicsRAM
X5670 @ 4.4/3.2GHz core/uncore, 1.36 vcore, 1.2... Gigabyte X58A-UD5 r2.0 w/FF3mod10 BIOS Sapphire Fury Nitro OC+ @ 1053/500, 1.225vGPU/1... 2x Samsung MV-3V4G3D/US @ 2000, 10-11-11-30-T1,... 
RAMHard DriveHard DriveHard Drive
1x Crucial BLT4G3D1608ET3LX0 @ 2000, 10-11-11-3... OCZ (Toshiba) Trion 150 120GB Hyundai Sapphire 120GB 3x Hitachi Deskstar 7k1000.C 1TB 
CoolingOSPowerCase
Noctua NH-D14 Windows 7 Pro x64 SP1 Antec TP-750 Fractal Design R5 
Audio
ASUS Xonar DS 
CPUMotherboardGraphicsRAM
i7-6800K @ 4.3/3.5GHz core/uncore, 1.36/1.2v ASRock X99 OC Formula (P3.10) GTX 780 (temporary) 4x4GiB Crucial DDR4-2400 @ 11-13-12-28-T2, 1.33v 
Hard DriveHard DriveCoolingOS
Intel 600p 256GB NVMe 2x HGST Travelstar 7k1000 1TB Corsair H55 (temporary) Windows Server 2016 Datacenter 
PowerCase
Seasonic SS-860XP2 Corsair Carbide Air 540 
  hide details  
Reply
post #59 of 66
Quote:
Originally Posted by biftek View Post
Where do do pull this **** from?
Higher failure rates? Ha! Look at the ballistix... To compare I've heard almost no stories of generic ram dying.
Better testing? How about this; Crucial, G.Skill, Corsair all make "generic" ram. You act like only off brand names make generic sticks.
Name your so called "cheap companies" please?

in fact some generic ram OC's just as good as the heat spreader counterparts
You know generic kingston ram has used the same D9's found in the old ballistix before, right?
Just because you had one non OCing generic stick doesn't mean they all suck.
Naked ram is the same as covered ram, you have to research to find ones that OC well.
I've had some generic A-DATA memory that OC'd great for what it was. 1000mhz at 5-5-4-12 to be exact.

i'd like to see your points backed up by facts.
I had 2 generic ram fail on me before
one ddr2-667 and one ddr-333

to the OP you picked an old GSkill kit with high price

when a fair compare between a ddr2-1066 cl5 kingmax or low reputation company memory for $45 or a ddr2-1066 cl5 gskill ram for $60, I will sure pick the GSkill for their support and quality
My PC
(23 items)
 
Home Server
(12 items)
 
Aorus X7 DT v7
(11 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 5820K ASUS X99 Deluxe NVIDIA MSI GTX1080 EK X NVIDIA MSI GTX1080 EK X 
RAMHard DriveHard DriveOptical Drive
32GB GSkill Trident Z DDR4 3000MHz CL14 Samsung 850 EVO 500GB SSD Seagate 4TB LG 14X BluRay Writer 
CoolingCoolingCoolingCooling
EK ASUS X99 Monoblock 2x EK XE480 Radiators Swiftech MCP35X (Dual DDC) Pump EK X3 250 Reservoir 
OSMonitorKeyboardPower
Windows 10 Pro X64 LG C6 55" UHD OLED Smart TV Logitech G710 Corsair RM1000 1000W 
CaseMouseMouse PadAudio
Corsair 900D Logitech G700S Razer ExactMat Pioneer VSX-930-K Receiver 
AudioAudioAudio
Pioneer S-RS77TB Speakers Pioneer S-RS3SW Subwoofer Logitech G35 
CPUMotherboardGraphicsRAM
Intel Core i7 7820HK Aorus HM175 NVIDIA GTX 1080 Notebook 32GB Kingston Hyper X DDR4 2400MHz 
Hard DriveHard DriveOSMonitor
512GB Samsung SM961 SSD Seagate FireCuda 2TB Windows 10 Pro 17.3" Sharp IGZO 4K 
KeyboardPowerCase
Aorus RGB Aorus 255W Power Adapter Aorus X7 v7 
  hide details  
Reply
My PC
(23 items)
 
Home Server
(12 items)
 
Aorus X7 DT v7
(11 items)
 
CPUMotherboardGraphicsGraphics
Intel Core i7 5820K ASUS X99 Deluxe NVIDIA MSI GTX1080 EK X NVIDIA MSI GTX1080 EK X 
RAMHard DriveHard DriveOptical Drive
32GB GSkill Trident Z DDR4 3000MHz CL14 Samsung 850 EVO 500GB SSD Seagate 4TB LG 14X BluRay Writer 
CoolingCoolingCoolingCooling
EK ASUS X99 Monoblock 2x EK XE480 Radiators Swiftech MCP35X (Dual DDC) Pump EK X3 250 Reservoir 
OSMonitorKeyboardPower
Windows 10 Pro X64 LG C6 55" UHD OLED Smart TV Logitech G710 Corsair RM1000 1000W 
CaseMouseMouse PadAudio
Corsair 900D Logitech G700S Razer ExactMat Pioneer VSX-930-K Receiver 
AudioAudioAudio
Pioneer S-RS77TB Speakers Pioneer S-RS3SW Subwoofer Logitech G35 
CPUMotherboardGraphicsRAM
Intel Core i7 7820HK Aorus HM175 NVIDIA GTX 1080 Notebook 32GB Kingston Hyper X DDR4 2400MHz 
Hard DriveHard DriveOSMonitor
512GB Samsung SM961 SSD Seagate FireCuda 2TB Windows 10 Pro 17.3" Sharp IGZO 4K 
KeyboardPowerCase
Aorus RGB Aorus 255W Power Adapter Aorus X7 v7 
  hide details  
Reply
post #60 of 66
Quote:
Originally Posted by Blameless View Post
Soild metal heatspreaders, as little as they typically do, are better than the OCZ design. If you are going to put a heatsink on something, you are doing so because air itself does not have the thermal conductivity to remove enough heat from as small an area as you want to cool. You do not want holes in your heatsink/heatspreader exposing the ICs/die to air. You want the most dense material with the highest thermal conductivity possible in contact with the ICs; then you want to give that material as much surface area as possible.

The OCZ heatspreaders are possibly the worst design of all. With the OCZ "heat spreaders" you have the ICs, thermal tape, and an open honeycomb.

-The total surface area of the heatspreaders is not significantly larger than that of the ICs (a problem with most heatspreaders), so even with zero thermal resistance and infinitly high thermal conductivity (both impossibilities in this case), you would, at best, have the exact same temps as with no heatspreaders at all.
-The ICs receive no cooling on most of their surface area, as the heatspreaders are full of holes.
- Even where there are holes, there is thermal tape, further increasing thermal resistance.
Excellent points but I disagree.

While I know this isn't the topic of the thread we all know that air is a terrible conductor of heat. It's more of an insulator than anything. Having all of that warm air trapped in between the heat spreader and the PCB isn't doing anything to help cooling because you are relying on low case temperatures and active cooling to help dissipate the heat and in the process are losing out on warm air's natural ability to rise up.

In either case the reason why heat spreaders are supposed to work is because you are increasing the thermally conductive surface area. Even a slight increase in surface area is better than nothing so with the OCZ heat spreaders you retain the air's ability to rise up and away from the modules as well as an increase in conductive surface area. This isn't just an OCZ fanboy trying to make a claim for his favorite memory.

It's the best of both worlds in my opinion.

Frankly I think heat spreaders are a complete sham unless you have extremely high ambient temperatures with active cooling and these companies like Crucial, G.Skill and yes, even OCZ with their Reaper series of memory are just out to make some extra cash. Let's face it, most of us won't put much more than 2.2v into our memory so slap some active cooling on some generic sticks without heat spreaders and you'll be just as well off.

With a fan pointed at them I doubt you'll see your temperatures rise above anything that the modules can handle and you can pocket the extra cash for something else.

The only reason I go with brand name modules is the warranty. Lifetime means peace of mind.
Original Nutta!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel i3 530 Gigabyte H55M-USB3 Sapphire 6850 Toxic Edition 2x2GB Corsair Dominator 
Hard DriveOptical DriveOSMonitor
60GB Muskin Callisto Deluxe, 640GB WD Caviar Black LG GH22NS50 22X DVD Writer Windows 7 Ultimate x64 LG W2242T 
KeyboardPowerCaseMouse
Razer Arctosa SilverStone Strider 750w Gold Corsair Graphite White 600T Cooler Master Storm Sentinel Advanced 
Mouse Pad
Thermaltake Dasher 
  hide details  
Reply
Original Nutta!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Intel i3 530 Gigabyte H55M-USB3 Sapphire 6850 Toxic Edition 2x2GB Corsair Dominator 
Hard DriveOptical DriveOSMonitor
60GB Muskin Callisto Deluxe, 640GB WD Caviar Black LG GH22NS50 22X DVD Writer Windows 7 Ultimate x64 LG W2242T 
KeyboardPowerCaseMouse
Razer Arctosa SilverStone Strider 750w Gold Corsair Graphite White 600T Cooler Master Storm Sentinel Advanced 
Mouse Pad
Thermaltake Dasher 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Memory
Overclock.net › Forums › Components › Memory › Memory: Why Pay More?