8Pack Binned 5ghz+ Delidded 7700K CPU's for sale at OCUK

Quick (probably noob) question, if I went for one of these chips, what is the degradation like on the liquid metal inside?

I'm hoping the answer is I would never need to ever open the CPU up again, clean and reapply at some point in the future :P
 
Quick (probably noob) question, if I went for one of these chips, what is the degradation like on the liquid metal inside?

I'm hoping the answer is I would never need to ever open the CPU up again, clean and reapply at some point in the future :P

Had my 4770k for a couple of years (was delidded and liquid metal) and never had an issue. Still going strong with its new owner afaik :)
 
Sorry to post this - but, these really arn't worth it.

I bought one of these from OCUK from the PoV that I wouldn't need to do it myself, and risk it.

While it does indeed hit 5Ghz (my particular sample @ 1.35vcore) - the delidding leaves much to be desired.

I use a Predator AIO - and my temperatures with this hit 81-85 in XTU.

I was curious as to why, and after speaking in other forums, seeing others results, I borrowed a friends Rockit delidding tool, and took a look. (Yes, I know this removes the "OCUK" warranty)

It seemed (and this is a guess as to the cause) that either intel's old glue hadn't been removed properly, or too much was reapplied - it was quite messy.

I reapplied my own Liquid Metal Ultra, relidded and have just tested.

Results:

OcUK delid: 81-85 on XTU stress @ 5Ghz 1.35 vcore

My delid: 59-61 on XTU stress @ 5Ghz 1.35 vcore.

I wasn't expecting such a huge difference, but there you go. Very disappointed in OCUK this time around.#

On reflection I should have taken some more pictures, but I didn't... those that I did = http://imgur.com/a/7MhP3
 
Last edited:
I checked afew of these myself and all was fine.

I will get them all checked now for consistency.

The product did perform exactly as expected though.

Good to hear dude, I'm still looking at pulling the trigger on one. I've got a retail on the way that I am going to check first but if I can't get what I want OC wise with that one going to get one.

One question, when you said 'the product did perform exactly as expected though' were you implying the OC of a stable 5ghz did?

I only ask it appears on this case the de-lidding aspect of this one didn't perform up to snuff from what you would expect to a de-lidded chip?

Many Thanks
 
I checked afew of these myself and all was fine.

I will get them all checked now for consistency.

The product did perform exactly as expected though.

These things happen - on the plus side, atleast I've done my own now.

Just had to make you guys aware to avoid others disappointment :)
 
OcUK delid: 81-85 on XTU stress @ 5Ghz 1.35 vcore

My delid: 59-61 on XTU stress @ 5Ghz 1.35 vcore.

I wasn't expecting such a huge difference, but there you go. Very disappointed in OCUK this time around.#

On reflection I should have taken some more pictures, but I didn't... those that I did = http://imgur.com/a/7MhP3

This is the reason, I will NOT buy it. OCUK should have re-checked everything before sold to the customers, it all hit and a miss!
 
Last edited by a moderator:
This is the reason, I will NOT buy it. OCUK should have re-checked everything before sold to the customers, it all hit and a miss!

Yeah but nowhere does it state you will actually get a drop in core temps with the delid. Yeah it's implied but that isn't actually the same. It would have been done and then binned and checked that it hits 5GHz and then packaged probably with no comment on the temps.

They may well see a large variation in temps because of this process and as long as it does what it states with the clock speed then they have done their job.

Yes it isn't right in that because it has been delidded etc then we should expect but unless they place a comment in the sales pitch of a guaranteed speed it really is just a bonus.
 
Yeah but nowhere does it state you will actually get a drop in core temps with the delid. Yeah it's implied but that isn't actually the same. It would have been done and then binned and checked that it hits 5GHz and then packaged probably with no comment on the temps.

They may well see a large variation in temps because of this process and as long as it does what it states with the clock speed then they have done their job.

Yes it isn't right in that because it has been delidded etc then we should expect but unless they place a comment in the sales pitch of a guaranteed speed it really is just a bonus.

Fair point.


To truly be on the safe side they should just not advertise the de-lidding with that thought process. The advertisement itself will invoke the expectation.

Well and truly take the point it is not stated as a guarantee.

Anyone could spend plenty of time debating wording both for and against, my thought is this one is more of an anomaly and as a whole OCUK is providing a service of a nice binned cpu and improved thermals of a delid.
 
Ok so got the replacement CPU fully lidded this time.

Testing is going well so far. Applied the settings in the 4.8ghz config thread and set to 50 multiplier at 1.350v.

Max temp with XTU and Heaven running was 65c with the NZXT x62 but normally around 60c.

Did try for 5.1 at 1.350v but no go so far.

All in I'm happy with the CPU and temps that I am getting:D
 
The binning was done before delid. All chips binned by me. Then technical team delided the CPU's and re-glued. I tested 5 at random after this.

All where completely fine as far as delid quality goes and running 5g at 1.35v or less.

Remember some chips are higher leakage.

I did a test today showing before and after delid temps on the same chip which I completed myself. Will post the results tomorrow as I left SSD at work.
 
The binning was done before delid. All chips binned by me. Then technical team delided the CPU's and re-glued. I tested 5 at random after this.

All where completely fine as far as delid quality goes and running 5g at 1.35v or less.

Remember some chips are higher leakage.

I did a test today showing before and after delid temps on the same chip which I completed myself. Will post the results tomorrow as I left SSD at work.

Oh really I thought it would have been binned after and then you could just sell the ones that don't cut it as Re-lidded as a middle ground product between stock and the 5GHz.

Cheers for the info though.

Look forward to seeing the results that you have from yesterdays tests. Just be good to see the averages if nothing else so it keeps peoples expectations and realisation that as you say there is some higher leakage in certain chips.
 
Yeah but nowhere does it state you will actually get a drop in core temps with the delid. Yeah it's implied but that isn't actually the same. It would have been done and then binned and checked that it hits 5GHz and then packaged probably with no comment on the temps.

That sounds like an answer from a politician.

It seems a delid wasn't properly carried out, and therefore temperatures were higher than the delid process should produce.

For the price premium (~£100), it should be cost effective (and common sense) to test each CPU after the delid process has taken place. It would take no more than about 10 minutes to pop a CPU into a test rig, fully load it and check that it hits it's target speed and that temps are decent.
 
Oh really I thought it would have been binned after and then you could just sell the ones that don't cut it as Re-lidded as a middle ground product between stock and the 5GHz.

Cheers for the info though.

Look forward to seeing the results that you have from yesterdays tests. Just be good to see the averages if nothing else so it keeps peoples expectations and realisation that as you say there is some higher leakage in certain chips.

As Ian has said, some chips will be higher leakage, and binning takes time. You want to be focusing on finding samples that will do 5Ghz within a certain voltage. That way also the customer benefits. This is because post de-lid, the CPU will likely require a certain degree less voltage due to reduced temps, meaning 5.1, or potentially even 5.2 may be possible depending on how thermally limited you are and how good the silicon is on your sample.
 
Honestly these are cheap as chips (pun intended) before the luxury of being able to buy these ready binned CPU's I personally have spent thousands of pounds trying to find a decent binned CPU.
To get a binned CPU for £100 extra is nothing the time it takes testing hundreds of them then delidding them then retesting, in comparison when you get a service on your car you get someone trained of which hundreds of thousands of people can do that job yet they charge you over £60 per hour. 8Pack is a one off only one of him in the World and he does all that for £84 don't forget the government gets £16 VAT for all his efforts.

I would have happily paid £250 extra it's a lot cheaper than doing it yourself unless you have the luck of the Irish.
 
That sounds like an answer from a politician.

It seems a delid wasn't properly carried out, and therefore temperatures were higher than the delid process should produce.

For the price premium (~£100), it should be cost effective (and common sense) to test each CPU after the delid process has taken place. It would take no more than about 10 minutes to pop a CPU into a test rig, fully load it and check that it hits it's target speed and that temps are decent.

They would be re-tested to the point that they still get the 5GHz clock as 8-Pack stated but the temp is still not relevant for their sales point in that they do not state anything about it.

And yeah it may seem that it is a politician answer but it is exactly what is being sold and thus deciding that just because you want it to be lower temp wise doesn't make it so :/
 
As Ian has said, some chips will be higher leakage, and binning takes time. You want to be focusing on finding samples that will do 5Ghz within a certain voltage. That way also the customer benefits. This is because post de-lid, the CPU will likely require a certain degree less voltage due to reduced temps, meaning 5.1, or potentially even 5.2 may be possible depending on how thermally limited you are and how good the silicon is on your sample.

Aye, in fairness I did think about that after I posted but at time was thinking actually taking a batch and de-lidding and then then testing means they could easily have 3 ranges:

- 4.6GHz at £385
- 4.8GHz at £415
- 5.0GHz at the £450

So at each price point the price goes up by £35. They all would have had a batch de-lidded prior to initial test and thus can be done quicker.

Then a quick test rules out the lower OC ones and then as more testing is done and thus time used the price goes up. That was all. But yes it works the other way with only cherry picking first off as well.

I mean I would be tempted to say that a lot of people may be tempted for a smaller overhead to pick up a relidded CPU for air cooling at the 4.6GHz for instance.
 
Back
Top Bottom