I'm sure that many forum members ( and not only ) saw my other A-Data memory tests. I'm not working for A-Data or something, I'm just picking their RAM because of nice price and performance 
This short test is about AX3U2000GB2G9-TG2 kit.
I wish to make max clock tests but for now I don't have good cpu for anything higher so maybe will add it in some time.
I will focus on 1600-2000 clocks mainly for 24/7 work and if anyone wish to see anything more then please send me pm.
I got it mainly because I had good deal , about 130$. I can't find it in US shops but 1st 2000-G version costs about the same in newegg and here it's about 15-20$ difference ( here, read in Poland
).
Press realise photo ( ~may 2010 ) is showing difference between 1st and 2nd version of AData 2000G.
some more important info:
Timings: 9-9-9-24 @ 2000
Voltage: 1.55-1.75V
Chips: PSC ( that's nothing new lately ) XEY627A3G-A ( that's new as I saw mainly XEE before )
No chips photo as I forgot to make it and thermal tape is kinda hard to remove so don't want to try it again.
SPD looks like this:
What more from producer:
"Thermal Conductive Technology (TCT) - Each memory chip is direct contact with the heat spreaders to perform immediate heat spreading from critical areas." Shortly ... "new" thermal tape that is really sticky and really have contact with all chips and higher heaspreader
... and more interesting High quality 2oz copper 8 layers PCB ... what can see from outside is that PCB is black
PC config:
i7 950 / water cooling / room temp ~22*C
Asus P6X58D-E, 0405 bios
Corsair HX850, 850W psu
Corsair F60 for system+2x PQI 32GB RAID0 for other data ( I don't need much space )
There is only 1 XMP profile 9-9-9-24 1.65V that isn't working on my board and can't even post.
Now some tests ( lowest stable voltage on each main timings ):
9-9-9-24 1T@2000
stock settings aren't interesting so I changed it a bit timings and voltage to 1.5V
8-9-8-24 1T@2000
I didn't expect 1.54V on these settings as older 2000G kit needed 1.58V+... but it's working
7-9-7-24 1T@2000
1.60V, I think it's quite nice while it's still under standard XMP voltage
6-9-6-24 1T@2000
Now something that took me some time to make stable on older 2000G kit and here no big problems 1.68-1.7V and 0.04-0.06V lower than older 2000G that is also on PSC chips.
6-8-6-24 1T@1800
Max stable@ 6-8-6-x, 1.65V. Here I was expecting more but on 1866 were single errors after ~30min. Memtest86+ is passing without problems up to 1866 but full load in windows is showing that max is about 1800 and maybe some MHz more but not much.
6-7-6-24 1T@1600
I didn't expect that this kit will make tRCD 7 on 1600 but yes it can, what more on 1.54V There aren't many PSC kits that can make it and it's my 1st kit with tRCD 7 on these clocks ( I tested a lot and different memory brands ).
9-9-9-24 1T@1600
Seems like nothing interesting can be done on these settings ... except voltage that is needed to make it stable - 1.2V and I wanted to try lower but just can't ...
Hard to say if it's better pcb or better chips but difference is big comparing to first revision of 2000G.
That's all for now. I have hope that my not best english won't be problem here. Also in this case there aren't many tests of A-Data 2000G v2.0 in the net so I have hope it will be helpfull at least for some of you who are looking for cheap and nice performing 3x2GB kit.
As I said before I wasn't able to make higher clock tests and I wish to try it as it's better than my older 2000G kit that made 7-10-7-24 1T@2250 and it was cpu limit.
Feel free to comment or ask questions here or on pm.
This short test is about AX3U2000GB2G9-TG2 kit.
I wish to make max clock tests but for now I don't have good cpu for anything higher so maybe will add it in some time.
I will focus on 1600-2000 clocks mainly for 24/7 work and if anyone wish to see anything more then please send me pm.
I got it mainly because I had good deal , about 130$. I can't find it in US shops but 1st 2000-G version costs about the same in newegg and here it's about 15-20$ difference ( here, read in Poland
).Press realise photo ( ~may 2010 ) is showing difference between 1st and 2nd version of AData 2000G.
some more important info:
Timings: 9-9-9-24 @ 2000
Voltage: 1.55-1.75V
Chips: PSC ( that's nothing new lately
) XEY627A3G-A ( that's new as I saw mainly XEE before )No chips photo as I forgot to make it and thermal tape is kinda hard to remove so don't want to try it again.
SPD looks like this:
What more from producer:
"Thermal Conductive Technology (TCT) - Each memory chip is direct contact with the heat spreaders to perform immediate heat spreading from critical areas." Shortly ... "new" thermal tape that is really sticky and really have contact with all chips and higher heaspreader
... and more interesting High quality 2oz copper 8 layers PCB ... what can see from outside is that PCB is black
PC config:
i7 950 / water cooling / room temp ~22*C
Asus P6X58D-E, 0405 bios
Corsair HX850, 850W psu
Corsair F60 for system+2x PQI 32GB RAID0 for other data ( I don't need much space
)There is only 1 XMP profile 9-9-9-24 1.65V that isn't working on my board and can't even post.
Now some tests ( lowest stable voltage on each main timings ):
9-9-9-24 1T@2000
stock settings aren't interesting so I changed it a bit timings and voltage to 1.5V
8-9-8-24 1T@2000
I didn't expect 1.54V on these settings as older 2000G kit needed 1.58V+... but it's working
7-9-7-24 1T@2000
1.60V, I think it's quite nice while it's still under standard XMP voltage
6-9-6-24 1T@2000
Now something that took me some time to make stable on older 2000G kit and here no big problems 1.68-1.7V and 0.04-0.06V lower than older 2000G that is also on PSC chips.
6-8-6-24 1T@1800
Max stable@ 6-8-6-x, 1.65V. Here I was expecting more but on 1866 were single errors after ~30min. Memtest86+ is passing without problems up to 1866 but full load in windows is showing that max is about 1800 and maybe some MHz more but not much.
6-7-6-24 1T@1600
I didn't expect that this kit will make tRCD 7 on 1600 but yes it can, what more on 1.54V
There aren't many PSC kits that can make it and it's my 1st kit with tRCD 7 on these clocks ( I tested a lot and different memory brands ).
9-9-9-24 1T@1600
Seems like nothing interesting can be done on these settings ... except voltage that is needed to make it stable - 1.2V and I wanted to try lower but just can't ...
Hard to say if it's better pcb or better chips but difference is big comparing to first revision of 2000G.
That's all for now. I have hope that my not best english won't be problem here. Also in this case there aren't many tests of A-Data 2000G v2.0 in the net so I have hope it will be helpfull at least for some of you who are looking for cheap and nice performing 3x2GB kit.
As I said before I wasn't able to make higher clock tests and I wish to try it as it's better than my older 2000G kit that made 7-10-7-24 1T@2250 and it was cpu limit.
Feel free to comment or ask questions here or on pm.
Last edited:
