collection for cgminer 7970 [Card as been sent! THANK YOU EVERYONE]

[jjiimm_64]

I am taking up a collection to send ckolivas a 7970.  it will be this one:
http://www.newegg.com/Product/Product.aspx?Item=N82E16814150585

http://www.newegg.com/Product/Product.aspx?Item=N82E16814102961

I am pledging 10 btc's to get things started.  will post a btc addy after things get rolling.
card:  560$
shipping  ~75$

btc addy:  18xk7bP8KeVm3XrL7p229dG1ntkEaWUUff
Jim

getting there.
thxs everyone:

jimbit	10	confirmed
graet	10	confirmed
anon	10	confirmed
Chefnet	7.58811018	confirmed
h4gb4s3	2	confirmed
gnar1ta$	5	confirmed
s..b....	10	confirmed
Pontius	0.41188982	confirmed
spiccioli	2	confirmed
cyberlync	3	confirmed
fydel	1.05	confirmed
miscreanity	1	confirmed
Isepick	10	confirmed
sveetsnelda	2.525	confirmed
xzion	2	confirmed
?	10	confirmed(not sure about this one.)
project10	2	confirmed
?	1	confirmed
outsidefactor	5	confirmed
sveetsnelda	10	confirmed
gnar1ta$	5	confirmed

total btc: 99.575

est conv: 5.85


  ______
total:  582.51375

cost:  568.5 + 60 = 628 total

close enough for me.  card has been sent.

Code:

Order Summary
Qty 	Product Description 	Price
Order #: 120577875(shipped via UPS Guaranteed 3 Day)
1   SAPPHIRE 21197-00-40G Radeon HD 7970 3GB 384-bit GDDR5 PCI Express 3.0 x16 HDCP Ready CrossFireX Support Video Card
    Item #: N82E16814102961

    VGA Standard Return Policy

    $559.99

Subtotal 	$559.99
Tax 	$0.00
UPS Guaranteed 3 Day 	$8.50
Order Total 	$568.49

Priority Mail International Parcels
   

Delivered
   

February 08, 2012, 7:25 am
   

AUSTRALIA

[1715467736]

1715467736

[1715467736]

1715467736

[1715467736]

1715467736

[1715467736]

1715467736

[1715467736]

1715467736

[Graet]

10btc from me too

awesome idea
onya jjiimm_64

[-ck]

I Heartily Endorse This Event Or Product

http://www.youtube.com/watch?v=lsiuxim8vsM

But seriously, wow, thanks! Though I'm partial to Sapphire cards, I guess I really shouldn't complain at all. These things cost 700+ dollars in .au where I am so I still wouldn't dream of buying one for myself.

http://www.pccasegear.com/index.php?main_page=index&cPath=193_1309

[jjiimm_64]

I already bought the xfx, but if the donations come in quick enough, I did put one of these in my cart.  I think newegg takes it out of avail if you have it in your cart. but I dont know for how long.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814102961


edit:  I was wrong, my cart was empty the next morning. sorry ck it will be the xfx unless more saphires come in.

[sveetsnelda]

Definitely stick with Sapphire if possible.  I know they're all reference boards, but many of the other brands are just becoming "cheap" (not to be confused with "inexpensive").  Sapphire is doing a damn good job right now and actually includes some accessories with the product (the XFX card came with almost nothing).  I'm trying to stick to Sapphire and MSI right now for the most part (unless I get a smokin' deal on something else).

[Chefnet]

add 7.58811018 Btc

[jjiimm_64]

I have this 10, but dont know who it is from

(
            [account] => ck7970
            [address] => 18xk7bP8KeVm3XrL7p229dG1ntkEaWUUff
            [category] => receive
            [amount] => 10
            [confirmations] => 17
            [txid] => 96bc82ec77d68f8712adeeb383a95169c0e9eb646534c77c14148ea1f79ed261
            [time] => 1327518111
        )

[Graet]

sent 10 a couple of hours ago
confirmed in irc

[Valalvax]

Holy shitballs, why is shipping so expensive? 75 dollars? Sure you don't mean pesos or something?

[sveetsnelda]

Something tells me that ckolivas doesn't live in the same country. 

[-ck]

US -> AU

[os2sam]

Something tells me that ckolivas doesn't live in the same country. 

Same country as what?

[gnar1ta$]

5 sent

[Pontius]

add 7.58811018 Btc

"Flattend" that amount a little. 

[jjiimm_64]

add 7.58811018 Btc

"Flattend" that amount a little. 

yes, which means I will probably eat the rounded amount since I rounded up.


Isn't there anyone else that want to be able to use cgminer with the 7970, and see it optimized (sorry ck if I am raising the expectations too much )

we are halfway there...  I have the card now.  Well, UPS said they delivered it. 

[Keninishna]

you could always just put the 7970 in a box and give ck remote access.

[jjiimm_64]

you could always just put the 7970 in a box and give ck remote access.

Don't you think many of us offered that. 

Don't you think he deserves at least this much. 

Can you imagine the number of hours he has put into the program that allows us to have hands free mining....??

[sveetsnelda]

you could always just put the 7970 in a box and give ck remote access.

You've apparently never done kernel debugging before. 

[Keninishna]

True I guess when trying to debug a graphics card, crashes are gonna happen and doing that remotely would suck.

[-ck]

True I guess when trying to debug a graphics card, crashes are gonna happen and doing that remotely would suck.

Indeed when I was working remotely on that 5970 card I literally had someone who would push the reset button for me 10,000 km away whenever he saw the screen go haywire telling him it had crashed, yet that wasn't entirely reliable either and I'd have to email him regularly... and all I was doing was adjusting fan control on it. We were in different timezones too so he really didn't appreciate me spinning the fans up and down while he was trying to sleep  

[rjk]

Man, if I had my PDU talking to my KVM properly, there wouldn't be any hassles at all. Give you a login to the KVM and PDU, and them you get remote console access including BIOS and remote reboot capability.

I need to figure out why they won't communicate with each other.

[sveetsnelda]

Same country as what?

The OP.

Nice OS/2 avatar, BTW. 

[jjiimm_64]

isn't there anyone else that wants to have the 7970 optimized by ck?

[-ck]

isn't there anyone else that wants to have the 7970 optimized by ck?

Perhaps people are wasted from donating to diablo for the same thing.

[Roadhog2k5]

Aren't you sending him the 7970 regardless if your donation quota is met or not? Or are you just going to keep it for yourself and the donations?

[sveetsnelda]

Asking him to cough up a few hundred on his own is a tall order.

Come on.  Let's get him a damn card.  Every donation after this point, I match 50 percent.

[gnar1ta$]

I can throw in some more but not until later in the week - getting Dwolla'd..wait...wait..wait some more.  Spent all my mining bitcoins on another rig.

[Pontius]

jimbit	10	confirmed
graet	10	confirmed
anon	10	confirmed
Chefnet	7.58811018	confirmed
unknown	2	confirmed
gnar1ta$	5	confirmed
anon	10
unknown	0.41188982	confirmed
unknown	2	confirmed

add 7.58811018 Btc

"Flattend" that amount a little. 

Looks like I haven't been precise enough. 
These tremendous 0.41188982 BTC were my contribution to round Chefnets 7.58811018.

[jjiimm_64]

jimbit	10	confirmed
graet	10	confirmed
anon	10	confirmed
Chefnet	7.58811018	confirmed
unknown	2	confirmed
gnar1ta$	5	confirmed
anon	10
unknown	0.41188982	confirmed
unknown	2	confirmed

add 7.58811018 Btc

"Flattend" that amount a little. 

Looks like I haven't been precise enough. 
These tremendous 0.41188982 BTC were my contribution to round Chefnets 7.58811018.

Nice.  I thought you meant I flattened it with my rounding.  or some would call it 'jimmy math'

[jjiimm_64]

Aren't you sending him the 7970 regardless if your donation quota is met or not? Or are you just going to keep it for yourself and the donations?

If the quota is not met, or I don't send the card, all donations will be returned.

[cyberlync]

Sent 3 BTC, don't have much.

edit: Unknown? My nick is right here, we even mine at the same pool

[jjiimm_64]

Sent 3 BTC, don't have much.

Thank you.   

[Isepick]

Sent 10.

1)ckolivas should be rewarded for his hard work by doing more hard work on the 7900 series

2)I fully intend to tear some 7990s up when they come out, and having cgminer already working on the 7970 cards would be nice.

[-ck]

 

Love you all so far. Thanks!

I'm pretty sure the existing dual card code new in cgminer 2.2+ will be easy to extend to 7990 once it's out.

[sveetsnelda]

I see:

3 BTC
1.05 BTC
1 BTC

50% = 2.525 BTC

Sent.

[Graet]

jjiimm_64 if you are still coming up short let me know, I will happily help cover any shortfall - I dont think any of us want our donations back

ckolivas joined Ozcoin in the early days of both cgminer and Ozcoin, I know first hand the effort he has put in to help ALL of us optimise our miners and earn the maximum number of Bitcoins/day our hardware is capable of.

dig deep folks this guy has spent countless hours helping us, this card will enable him to volunteer more hours and help the community more
every 0.01Btc helps, for those that can afford more - well that helps more

[xzion]

2 from me.

[spiccioli]

2 from me.

I've sent 2 BTCs a couple days ago.

spiccioli

[jjiimm_64]

Thank you everyone.  keep them coming we are almost there.  I will lock the post when I receive enough.

[outsidefactor]

I just sent 5 BTC.

Let's help make cgminer even better!

[sveetsnelda]

I see another 20BTC since I last reconciled.  As soon as my next transaction clears, I'll send the 10BTC to match it.

[jjiimm_64]

wow,  thank you everyone. will update the list soon.

[sveetsnelda]

Since my last update, I see:

2 BTC
10 BTC
2 BTC
1 BTC
5 BTC

Total = 20 BTC

50 percent = 10 BTC

10 BTC Sent.

[kano]

Hmm I thought the first post said 100 BTC was the amount required ...

[sveetsnelda]

I believe 100 BTC at the time of the first post would have been around 650-660.  BTC/USD has dropped/fluctuated a bit since then...

[gnar1ta$]

Sent 5 more 

[jjiimm_64]

NICE..  calculating now.......

[cyberlync]

total btc: 99.575

est conv: 5.85


  ______
total:  582.51375

cost:  568.5 + 60 = 628 total

close enough for me.  card has been sent.

Great news!

[-ck]

Awesome awesome awesome 

As you may have seen I've done some preliminary work to the git tree since this collection started anyway so that cgminer will run at least satisfactorily on 7970 if not optimally just yet.

Thanks to everyone who donated 

[-ck]

Thanks very much everyone. Arrived, shiny and all

Took a lot of fluffing around to get it going thanks to AMD fail. Had to make sure the 7970 was in the first slot and manually edit xorg to make it work. The older driver detected the 6970s only and the newer experimental 79xx driver only detected the 7970. Then I did a little overclocking experimenting and found a nice sweetspot at 1200/1050 (it has a -150 gpu-memdiff limit).

See if you can spot the 7970 

Code:

cgminer version 2.2.3 - Started: [2012-02-09 02:30:05]
--------------------------------------------------------------------------------
 (5s):2086.3 (avg):1968.1 Mh/s | Q:286  A:0  R:0  HW:0  E:0%  U:0.00/m
 TQ: 5  ST: 5  SS: 0  DW: 5  NB: 3  LW: 0  GF: 0  RF: 0
 
 Block: 0000013f08b93da9e177e5543ffe6834...  Started: [02:35:58]
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0:  74.0C 4473RPM | 685.2/683.4Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 1:  74.0C 5390RPM | 427.9/426.4Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 2:  73.5C 4663RPM | 428.1/424.4Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 3:  73.0C 4187RPM | 437.6/436.0Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
--------------------------------------------------------------------------------

This is with an unmodified 2.2.3 with its defaults so that's what you can expect from cgminer at this stage at those clock speeds till I can wedge in a faster kernel.

[jjiimm_64]

Thanks very much everyone. Arrived, shiny and all

Took a lot of fluffing around to get it going thanks to AMD fail. Had to make sure the 7970 was in the first slot and manually edit xorg to make it work. The older driver detected the 6970s only and the newer experimental 79xx driver only detected the 7970. Then I did a little overclocking experimenting and found a nice sweetspot at 1200/1050 (it has a -150 gpu-memdiff limit).

See if you can spot the 7970 

Code:

cgminer version 2.2.3 - Started: [2012-02-09 02:30:05]
--------------------------------------------------------------------------------
 (5s):2086.3 (avg):1968.1 Mh/s | Q:286  A:0  R:0  HW:0  E:0%  U:0.00/m
 TQ: 5  ST: 5  SS: 0  DW: 5  NB: 3  LW: 0  GF: 0  RF: 0
 
 Block: 0000013f08b93da9e177e5543ffe6834...  Started: [02:35:58]
--------------------------------------------------------------------------------
 [P]ool management [G]PU management [S]ettings [D]isplay options [Q]uit
 GPU 0:  74.0C 4473RPM | 685.2/683.4Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 1:  74.0C 5390RPM | 427.9/426.4Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 2:  73.5C 4663RPM | 428.1/424.4Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
 GPU 3:  73.0C 4187RPM | 437.6/436.0Mh/s | A:0 R:0 HW:0 U:0.00/m I: 9
--------------------------------------------------------------------------------

This is with an unmodified 2.2.3 with its defaults so that's what you can expect from cgminer at this stage at those clock speeds till I can wedge in a faster kernel.

Ck:  can you do the comparisons with the clock not set so high.  at 1200 they run pretty hot with a high fan, also if you just do down about 50 on the clock, you can reduce the voltage down to 1120 or so.

I am currently running   at volt: 1100  clock: 1100   650 Mhash diablo

[-ck]

Hmm I usually run my cards flat out, and I'm pretty sure the current cgminer kernel being 9 months old isn't going to remotely come near the diablo speeds but I can have a play... brb

[-ck]

Ck:  can you do the comparisons with the clock not set so high.  at 1200 they run pretty hot with a high fan, also if you just do down about 50 on the clock, you can reduce the voltage down to 1120 or so.

I am currently running   at volt: 1100  clock: 1100   650 Mhash diablo

At your clock settings of 1100 (+5% powertune) I'm getting 638 Mhash. Voltage doesn't change on linux with ADL it seems.
Doing:

Code:

export GPU_USE_SYNC_OBJECTS=1

was mandatory for getting rid of the CPU usage bug. Interestingly with that feature enabled, there is an inflexion point in intensity when the CPU usage jumps up dramatically. With the cgminer default of 2 GPU threads, it is low CPU usage up to intensity 11. With 1 GPU thread, it is low CPU usage up to intensity 13. Both seem to provide only about 1MH difference.

[-ck]

And just for completeness, after the intensity investigation, I bumped it back up to the 1200/1050(+5%) clockspeed and put intensity up to 11 which seems the sweet spot. The final performance after it stabilised is:

Code:

GPU 0:  72.0C 4324RPM | 695.1/695.3Mh/s | A:98 R:1 HW:0 U: 9.79/m I:11

[jjiimm_64]

Looks like I will have to figure out how to get this thing working on linuxcoin...

I imagine it is something like

curl somespecialfile
edit xorg.conf directly

aticonfig initialize

?

[-ck]

Looks like I will have to figure out how to get this thing working on linuxcoin...

I imagine it is something like

curl somespecialfile
edit xorg.conf directly

aticonfig initialize

?

See my post here:
https://bitcointalk.org/index.php?topic=28402.msg738468#msg738468

That has the basics.

Feel free to ask about anything else. After the effort you put into getting the funds together and card to me, you deserve all the software support you need. As does sveetsnelda for his excellent generosity and everyone else who donated.

[outsidefactor]

Many thanks to ck the hard work and everyone else who donated!

I just changed over to 2.2.3 on my windows machine with two shiny new Sapphire 7970s and am getting 513 and 537 at stock stock speeds (running at a leisurely 78 degrees), a huge jump over 2.2.1.

Many thanks!

[-ck]

Many thanks to ck the hard work and everyone else who donated!

I just changed over to 2.2.3 on my windows machine with two shiny new Sapphire 7970s and am getting 513 and 537 at stock stock speeds (running at a leisurely 78 degrees), a huge jump over 2.2.1.

Many thanks!

Excellent. Don't forget to use intensity 11 if you can.

[outsidefactor]

Many thanks to ck the hard work and everyone else who donated!

I just changed over to 2.2.3 on my windows machine with two shiny new Sapphire 7970s and am getting 513 and 537 at stock stock speeds (running at a leisurely 78 degrees), a huge jump over 2.2.1.

Many thanks!

Excellent. Don't forget to use intensity 11 if you can.

I use dynamic on my main card and have been using 14 on the second as it's a Crossfire only card (I haven't progressed to the whole eyefinity 2 x 6 thang yet ). I've changed to 11 and it seems to have basically the same rate. Coolness!

[-ck]

Working on the 7970 tuning, I have tried to port both the diapolo and diablo kernels to cgminer. Alas neither of them are actually working yet, so instead I started modifying the existing poclbm kernel in cgminer to improve throughput. This should work on other GPUs as well as the 7970, but I have no idea if it's better or worse than phatk.

When it's released it will get a new date/version number, but I haven't changed the number right now so that people can download it now and give it a try:
https://raw.github.com/ckolivas/cgminer/kernels/poclbm120203.cl

Remember to delete any .bin files and if you're not on a 7970 with the latest cgminer, you'll have to tell it to use that kernel with -k poclbm.

So what's the damage? Well on the 7970 at 1200/1050 clocks, which was getting 694MHash, it's now getting 711Mhash. The 7970 has this unusual behaviour where the hashrate slowly rises for the first 5-10 minutes.

[d3m0n1q_733rz]

Working on the 7970 tuning, I have tried to port both the diapolo and diablo kernels to cgminer. Alas neither of them are actually working yet, so instead I started modifying the existing poclbm kernel in cgminer to improve throughput. This should work on other GPUs as well as the 7970, but I have no idea if it's better or worse than phatk.

When it's released it will get a new date/version number, but I haven't changed the number right now so that people can download it now and give it a try:
https://raw.github.com/ckolivas/cgminer/kernels/poclbm120203.cl

Remember to delete any .bin files and if you're not on a 7970 with the latest cgminer, you'll have to tell it to use that kernel with -k poclbm.

So what's the damage? Well on the 7970 at 1200/1050 clocks, which was getting 694MHash, it's now getting 711Mhash. The 7970 has this unusual behaviour where the hashrate slowly rises for the first 5-10 minutes.

The problem, I believe, is that the GCN cards are built to take large vector counts and perform a single instruction on them all at once.  This is in contrast to the small vector count of VLIW which can perform large instructions quickly.  So, a straight-forward 16-vector miner with simple instructions should work better than a 4-vector miner with multiple instructions.  Granted, this is speculation, but from what I've seen of the hardware specs, it should hold true.

[-ck]

New release of cgminer 2.2.5 with fresh kernel and no more zero binary error, coping with multiple different cards at last and working well with sdk 2.6.

7970 running at 1200/1050+5% is getting 714 Mhash with -I 11 and cgminer 2.2.5 defaults.

[-ck]

The problem, I believe, is that the GCN cards are built to take large vector counts and perform a single instruction on them all at once.  This is in contrast to the small vector count of VLIW which can perform large instructions quickly.  So, a straight-forward 16-vector miner with simple instructions should work better than a 4-vector miner with multiple instructions.  Granted, this is speculation, but from what I've seen of the hardware specs, it should hold true.

I tried it, and a 16 vector simplest possible mining kernel had performance which was, unfortunately, appalling. GCN with SDK 2.6 (the only one it works with) really does not want any vectors at all.

[d3m0n1q_733rz]

The problem, I believe, is that the GCN cards are built to take large vector counts and perform a single instruction on them all at once.  This is in contrast to the small vector count of VLIW which can perform large instructions quickly.  So, a straight-forward 16-vector miner with simple instructions should work better than a 4-vector miner with multiple instructions.  Granted, this is speculation, but from what I've seen of the hardware specs, it should hold true.

I tried it, and a 16 vector simplest possible mining kernel had performance which was, unfortunately, appalling. GCN with SDK 2.6 (the only one it works with) really does not want any vectors at all.

Hmm, interesting.  Did you drop the worksize in your tests as well?

[-ck]

The problem, I believe, is that the GCN cards are built to take large vector counts and perform a single instruction on them all at once.  This is in contrast to the small vector count of VLIW which can perform large instructions quickly.  So, a straight-forward 16-vector miner with simple instructions should work better than a 4-vector miner with multiple instructions.  Granted, this is speculation, but from what I've seen of the hardware specs, it should hold true.

I tried it, and a 16 vector simplest possible mining kernel had performance which was, unfortunately, appalling. GCN with SDK 2.6 (the only one it works with) really does not want any vectors at all.

Hmm, interesting.  Did you drop the worksize in your tests as well?

Absolutely. I tried all sorts of combinations. Specifically the optimum, as always, is using the card's reported max_work_size and then dividing that by the vector size. It gave the least worst performance... but we're talking 20% of the performance of running no vectors.

[d3m0n1q_733rz]

The problem, I believe, is that the GCN cards are built to take large vector counts and perform a single instruction on them all at once.  This is in contrast to the small vector count of VLIW which can perform large instructions quickly.  So, a straight-forward 16-vector miner with simple instructions should work better than a 4-vector miner with multiple instructions.  Granted, this is speculation, but from what I've seen of the hardware specs, it should hold true.

I tried it, and a 16 vector simplest possible mining kernel had performance which was, unfortunately, appalling. GCN with SDK 2.6 (the only one it works with) really does not want any vectors at all.

Hmm, interesting.  Did you drop the worksize in your tests as well?

Absolutely. I tried all sorts of combinations. Specifically the optimum, as always, is using the card's reported max_work_size and then dividing that by the vector size. It gave the least worst performance... but we're talking 20% of the performance of running no vectors.

Weird.  I would have thought it to do better considering the 16-vector ALUs.  VLIW actually showed the best output for me so long as I used VECTORS8 and GOFFSET=false as I'm using an HD5450.  But that comes with new architectures I suppose.  I wish I could find more literature for programming for GCN, but it's so new that I can't find much.  Combine that with not being able to test the programming, and I might as well stick with modifying Phatk2 (which is taking a while considering life's little disruptions lately).  I'll save-up for a 7970 and try to figure out what I can do about GCN code.  I'm thinking about having the code alternate between two variables in case it has any read/write timing conflicts.  On the other hand, it probably won't do much more than add another GPR and a few more ALUs.  Considering its instruction execution timing, it probably won't matter.  Too many theories to be pondering at the same time.  I'll rest on it and see what I can come up with.
If you have any documentation, I would really appreciate it.

[-ck]

cgminer 2.2.6 out. 2+ more mhash on 7970. It's getting harder and harder to extract much more

1200/1050+5% clocks, intensity 11 - 717 Mhash.

[d3m0n1q_733rz]

cgminer 2.2.6 out. 2+ more mhash on 7970. It's getting harder and harder to extract much more

1200/1050+5% clocks, intensity 11 - 717 Mhash.

I'm going to have to look at what methods you're using.  I'm curious as to how the programming differs between VLIW and GCN.

[-ck]

cgminer 2.2.6 out. 2+ more mhash on 7970. It's getting harder and harder to extract much more

1200/1050+5% clocks, intensity 11 - 717 Mhash.

I'm going to have to look at what methods you're using.  I'm curious as to how the programming differs between VLIW and GCN.

Direct link to the kernel in the git tree:
https://github.com/ckolivas/cgminer/blob/master/poclbm120214.cl

It's using a worksize of 256, 1 vector (i.e. no vectors) and intensity 11.

[-ck]

Cgminer 2.2.7 out. This version fixes the bug with 12.2 ATI drivers. Reworked to use -w 64 by default on Tahiti which is worth just under 1 more MHash. 718.5 MHash now at 1200/1050+5% clocks intensity 11.

[-ck]

cgminer 2.3.2 is out with a new poclbm kernel I've been bashing with a mallet for a week to try and extract some more out of it, and I hit my target which is 720 MHash at 1200/1050+5% clocks intensity 11 with 12.3 amd drivers.  

[jjiimm_64]

you da man....


I am stuck on windows with untill I can get linuxcoin to cooperate.

at   1125/1000  I11     672Mh   I am happy with that..  I have not tried to clock it up..  think I will try that now.