Diamond Viper Radeon HD 3650 1GB

Article Index:   

HOW WE CONFIGURED THE TEST SYSTEMS: We tested all of the graphics cards used in this article on an MSI P6N Diamond motherboard powered by a Core 2 Duo E6550 processor and 2GB of low-latency Corsair RAM. The first thing we did when configuring our test system was to enter into the BIOS and set all values to their "optimized" or "performance" default settings. Then we manually configured the memory timings and disabled any integrated peripherals that wouldn't be put to use. The hard drive was then formatted, and Windows Vista Ultimate was installed. When the installation was complete we fully updated the OS, and installed the latest DX10 redist and various hotfixes along with the necessary drivers and applications.

HotHardware's Test System
Viewing the Ultimate Vista


Processor -

Motherboard -




Video Cards -



Memory -


Audio -

Hard Drive -

HARDWARE USED:

Intel Core 2 Duo E6550

MSI P6N Diamond
nForce680i SLI Chipset

Diamond Radeon HD 3650 1GB
Sapphire Radeon HD 3650 512MB
Sapphire Radeon HD 3850 512MB
Gigabyte GeForce 8600 GTS
NVIDIA GeForce 8800 GT

2GB Corsair Dominator DDR2
2x1GB, 4-4-4-12 (1T)

Integrated Creative X-Fi

Seagate Barracuda 7200.10
750GB - 7200rpm - SATAII



OS -

Direct X -

Video Drivers -




Synthetic (DX) -
DirectX 9 -
DirectX 10 -
OpenGL -
DirectX 10 -
DirectX 10 -
DirectX 10 -
 

RELEVANT SOFTWARE:

Windows Vista Ultimate

DirectX 10

NVIDIA Forceware v169.25
ATI Catalyst v8.3

BENCHMARKS USED:

3DMark06 v1.1.0
Half-Life 2: Episode 2
Company of Heroes
ET: Quake Wars 1.4
Crysis Single-Player Demo
Bioshock 1.1
PT Boats: Knights of the Sea



Performance Comparisons with 3DMark06 v1.1.0
Details: www.futuremark.com/products/3dmark06



3DMark06
3DMark06 is the most recent addition to the 3DMark franchise. This version differs from 3Dmark05 in a number of ways, and includes not only Shader Model 2.0 tests, but Shader Model 3.0 and HDR tests as well. Some of the assets from 3DMark05 have been re-used, but the scenes are now rendered with much more geometric detail and the shader complexity is vastly increased as well. 3DMark06 also employs much more lighting and there is extensive use of soft shadows. With 3DMark06, Futuremark has also updated how the final score is tabulated. In this latest version of the benchmark, SM 2.0 and HDR / SM3.0 tests are weighted and the CPU score is factored into the final tally as well.





Diamond's HD 3650 finds itself with the short end of the stick in the 3DMark06 testing with by far the lowest scores of any of the cards used for this review.  The real crux here is that Diamond's version of the card comes with double the memory buffer than Sapphire's model, yet the slower DDR2 memory speeds employed cause it to lose any advantage whatsoever.  Instead, it winds up being nearly 30% slower as shown in the overall results graph.   

Tags:  Radeon, HD, diamond, Radeon HD, GB, Viper, IP, AM

Image gallery

Related content

Comments

Comments
Crisis Causer 6 years ago

Oh god... it's exactly what I expected.  I seem to recall specifically laughing at the concept of 1GB 3650's in another thread here.  It's a silly marketting gimmick and it also shows why it's important to avoid anything slower than GDDR3 memory.  Aye caramba.  Stuff like this actually annoys me.  Words cannot quite describe the feeling... disappointment with some anger.  It's almost bloated, but it's more like it's just a waste of such potential.

ice_73 6 years ago

 it exceeded my expectations (negativly) too much memory on a slow gpu decrease performance further ( like the radeon 9800). although honestly, it could be the drivers that limit the speed on this card, but if it is i dont see performance improve so much to make this card take the lead. nice to see diamond back though but this card isnt the beast i was expecting it to be. 

Bitbrain 6 years ago

This is just like the MHz race of years past. All this is for are for the people who only know how to turn a computer on a run MS Word and iTunes. Just like in the MHz wars all these people know is a bigger number is better so to market to these people they give them a bigger number to oogle at. And just like in the MHz wars these people don't know the drawbacks of haveing just a bigger number without factoring in other variables.

Next up we will sell you a Geo Metro with 50" tires.....bigger number is better. But if enough people buy the card because it has a bigger number on the box and makes it profitable to do so then I guess thats good marketing. For those of us "in the know" we know better than to judge a book by it's bigger number.

Super Dave 6 years ago
That was a very interesting review...one of the rare cases where throwing more memory at it failed to speed things up. Heck, it even slowed things down! 
ice_73 6 years ago

[quote user="Super Dave"]That was a very interesting review...one of the rare cases where throwing more memory at it failed to speed things up. Heck, it even slowed things down!  [/quote]

not to argue or flame, but this isnt really that rare, alot of times if a thrid party manufacterer goes and adds more memory than recomended the board will see no benifit, or decrease in speed. of course though if a manufacterer makes a 256 and 512 version the 512 *might* be faster but its 99.9% never faster when going over the manufacterer's rating... (i hope this post makes sense... ) 

Crisis Causer 6 years ago

It is one hell of a marketting thing though.  I bet if they had 1.5GB 8800GTX's, people would have paid an even higher premium for them.  Yet that extra memory would never have been used.  I only hope that people do some simple research on the interweb before purchasing.  It really is as easy as searching for "Diamond Radeon 3850 1GB review" on Google.  I spend days and hours just going through dozens of reviews before I commit to a new video card, but people should at least go through one.

ice_73 6 years ago

[quote user="Crisis Causer"]

It is one hell of a marketting thing though.  I bet if they had 1.5GB 8800GTX's, people would have paid an even higher premium for them.  Yet that extra memory would never have been used.  I only hope that people do some simple research on the interweb before purchasing.  It really is as easy as searching for "Diamond Radeon 3850 1GB review" on Google.  I spend days and hours just going through dozens of reviews before I commit to a new video card, but people should at least go through one.

[/quote]

i completly agree, i literally spends days and hours looking at reviews (even though i dont have money to buy anything ) and before i buy anything over 100, i make sure it is the best thing i can get for my money.
Post a Comment
or Register to comment