OC on Future Reviews POLL

Do you want to see OC results in Initial Architecture Reviews & not in Follow ups

  • Total voters
    72
L

Lord_Exodia

Supreme [H]ardness
Joined
Sep 29, 2005
Messages
7,008
Hypothetically speaking IF [H]ardOcp is reviewing and evaluating a card right now that Might come out soon, I really hope that on a new architechture that the team here at the [H] would overclock that card to see what this Hypothetical card could perform overclocked. This information may be critical to come to a final purchase decision.

I would like this data on the initial article (right after NDA expires) and not on a follow up article. :)

Now I can't remember paying membership fees here so I'll take what I can get but I'm just throwing this out there just to see how people feel.

So what say you people? Agree, Disagree, Don't Care, Work the reviewers until they Die of Insomnia.
 
I disagree, id rather have the new tech review out as quick as possible, and let the OC come out later when theyve had time to play with it.

OCs are never guaranteed on any hardware, therefore you should not be relying on it for your purchasing decisions.
 
Not saying your viewpoint is wrong The Mac but what if you are overclocking your current high end card and it outperforms the stock performane by a slight margin of the new card and because of that fact your not sure if it's worth a buy or not.

Say that new card is Super overclockable because of the new architectures power efficiency etc.. and it can overclock to unlock insane performance improvements and when the follow up comes out the new card is sold out everywhere.

Just saying LOL
 
I agree its always good to see the fully potential of a card in every review.. some cards are really sensitive to overclocking and the difference can be big from stock to a overclocked state.. anyway [H] only test a couple of games for review and the overclocked card should not add excessive work for 3 or 4 games.. =).
 
Lord_Exodia said:
Not saying your viewpoint is wrong The Mac but what if you are overclocking your current high end card and it outperforms the stock performane by a slight margin of the new card and because of that fact your not sure if it's worth a buy or not.

Say that new card is Super overclockable because of the new architectures power efficiency etc.. and it can overclock to unlock insane performance improvements and when the follow up comes out the new card is sold out everywhere.

Just saying LOL
Click to expand...

I agree to some extent.

But id rather have the initial review out ASAP.

I want to see the new tech, OCs are ancillary.

Besides, you never know if the review sample was cherry-picked, or they hit the silicon lottery, etc

You wont really know till they show up in the wild and see what people are averaging.
 
The Mac said:
I disagree, id rather have the new tech review out as quick as possible, and let the OC come out later when theyve had time to play with it.

OCs are never guaranteed on any hardware, therefore you should not be relying on it for your purchasing decisions.
Click to expand...

This

For GPUs I always make my decision on stock performance. I threat OC as an added bonus, I don't want to be left crawling on the sidewalk if I get a product that's a bad overclocker.
 
The Mac said:
I agree to some extent.

But id rather have the initial review out ASAP.

I want to see the new tech, OCs are ancillary.

Besides, you never know if the review sample was cherry-picked, or they hit the silicon lottery, etc

You wont really know till they show up in the wild and see what people are averaging.
Click to expand...
I disagree because ^this.
 
Lord_Exodia said:
Not saying your viewpoint is wrong The Mac but what if you are overclocking your current high end card and it outperforms the stock performane by a slight margin of the new card and because of that fact your not sure if it's worth a buy or not.

Say that new card is Super overclockable because of the new architectures power efficiency etc.. and it can overclock to unlock insane performance improvements and when the follow up comes out the new card is sold out everywhere.

Just saying LOL
Click to expand...

There are too many hypotheticals there, I haven't even used my last few cards OCed, the benefits were so small. And if a new hardware is coming out that'll be suprer OCable (highly unlikely as cards pushing 250-300W) the company that brings it out will make as much noise about it as possible.
 
The Mac said:
I disagree, id rather have the new tech review out as quick as possible, and let the OC come out later when theyve had time to play with it.

OCs are never guaranteed on any hardware, therefore you should not be relying on it for your purchasing decisions.
Click to expand...


Pretty much this. I actually prefer smaller articles separating the base card vs the OC'ed version. Some of these reviews get too fricken big to the point they aren't worth reading besides looking at the graphs and the information starts to overlap.

I am also in the boat of an OC being an added bonus, not one that should be expected. Even the hypothetical high OCing ability, which I assume was referenced to the 750/750 Ti, doesn't always mean big improvements. Those cards overclocked like crazy, but were held back by other means and still didn't come close to a 660 performance for a few bucks more. Going off topic, but as M76 already mentioned, there are just too many hypotheticals.
 
A lot of the decision comes down to time factor. We rarely have butt loads of time for evaluation of new release products. We have to be choosy on what is tested and written. We also do not have a team of people reviewing each video card, we individually test video cards, it is a one man operation from start to finish, all testing, and writing. I often go without sleep just to get it done on time.

When overclocking, time must be spent to find the stable overclock. You cannot do rushed jobs on overclocking, it takes time to hone in on the right gpu/memory frequency, it takes time to tweak, it takes time to burn in and test the overclock and really make sure it is stable for prolonged use. What good is reporting an overclock that only lasts 30 minutes? No, we need to test for days to make sure its stable, else it means nothing. We have to do this not just for the card being reviewed, but also the comparison cards, each one takes time to get it right and make good comparisons. That level of testing and writing is a full review in itself.

I'm sure this level of attention and detail is appreciated. We don't rush stuff out just to have it out, like some other websites. When we report things, it is well tested and solid, and you can count on that data. I am not going to degrade quality to rush things out. This will not change. Period.
 
Last edited:
This is why [H] is my go-to place for hardware reviews. So many places still do canned benchmarks because they're easy to run and allows them to race to get out a "review" first. [H] reviews are always detailed with a lot of thought put into the information given at a typical user-level, and I really appreciate the time and care put into them.
 
i don't get it why so many other sites jumped so eagerly on the OC band wagon

Wouldn't consumers expect performance out of the box and not with possibly breaking the card warranty increasing its voltage ?

as someone said OC is a nice bonus but i would never buy a product based just on its OC potential
 
Brent I have been coming here for years and you know me enough by now to know that I appreciate and understand all the hard work you guys do. :)

My thing is simply asking for some overclocking performance get fit in the review.

Sometimes some things get thrown in the review that I don't feel influence my buying decision. If something else has to give in order to get a chance to overclock just to give us a rough idea of what it can do then I'm all for it.

If instead of testing 7 games you guys test the 5 most taxing and that allows time to show some OC performance then that's cool.

I don't think we need to see 10-12 games since there are 30 other reviews that can give us a rough idea of what is up. Yours is the one that counts for me and many others so getting it just right is important to me.

For me, Especially when a new architecture is concerned. Overclocking is VERY important. For me (and I believe some others) knowing how high the Voltage goes with the OC software, how far you can get it core/memory and how hot it gets etc.. is more important than getting OC numbers for every game in your review.

I guess I'd simply say that I feel OC results of some kind should be fit in the review for brand new architecture releases (hypothetically speaking if one were to come out soon) :D

My perfect review for a hypothetical new gen release would be like this
_______________________
Top 5 taxing games

New card vs Top 2 old cards e.g. Reference 780Ti & Reference 290X max playable settings

You can skip apples to apples unless you have time

Overclocking results on the card (Max, Core/Mem)

Temp and Noise Before/After OC

Show performance in the most taxing game or 2 before and after OC

Your thoughts on cards potential and if a award is deserved.
------------------------------------
 
Proper overclock testing takes more time than you realize.

We won't be changing our format or quality, thanks for the feedback.

Overclocking will be dealt with properly.
 
I can never get the OCs that review sites seem to get so I'd rather just get a regular review out ASAP. :(
 
I disagree completely with the OP. OC is so random, There is no guarantee that any card will overclock much at all. IF you are basing your purchase on what it might overclock to then you are in for disappointment. It also has led to a lot of people returning cards because they haven't been great overclockers, which I think is entirely wrong.
 
wabbitseason said:
Every review should be comparing stock-to-stock and maxOC-to-maxOC.
Click to expand...

I'd like to amend this from "maxOC" to "moderate, typically-achievable OC". It would be at [H]'s discretion to define what constitutes a "typical overlock" for every card.
 
I agreed, but with a caveat: if it delays the review of the product, don't do it. Otherwise, I'd love to see it in the original review.
 
Don't forget that reviewers often don't have the item to be reviewed for very long before the release date and there's only so much that can be done in that time. Overclocking in particular can be very time-intensive. An initial overclock can seem stable, but fail after a few hours or days and reviewers often don't have that time available.
 
After initially voting for yes I think I'd greatly prefer to see some time spent on benches for >60 Hz monitors instead instead of overclocking.
 
Lord_Exodia said:
Fair enough. How much later is acceptable to you?
Click to expand...

As long as it takes.

If you're that guy that can't wait after something new comes out, then you must also accept the risks involved. That is not knowing how much OC is typically possible.

OC has many variables, it's best to wait for as many tests as possible if you're interested in that. As one card can never be representative of the OC potential of a product. And basing your purchases on a single OC review can lead you far astray. Samples can be hand picked. Or if they get their sample from retail sources they can get a very bad example which is not representative of the average potential.
 
I could see several issues with OC results on reviews for new equipment:

1. If these are engineering/review samples (or in any other way not quite a finished product, abused by a prior review site, etc), then the OC results are almost meaningless and a potential waste of time for the reviewers.

2. The manufacturers typically develop the enthusiast platform equipment based on the GPU manufacturer's chips, so there is the potential to see a lot of variance with custom cards. Admittedly, this is less likely with re-branded vanilla designs, so I personally would rather know whose overclockable card I should buy if I'm in the market for one, not necessarily whether the architecture is a good OC candidate or not.

3. A proper OC takes time, potentially lots of it. It also takes feedback from a community of enthusiasts based on their own experiences mixed with some trial and error. I don't know that the kind of review time this would take would necessarily be cost effective (after all there is a business aspect to [H] ).
 
So despite everything said so far in this thread. Who wished there were some overclocking information in the Review done here?

I would have preferred even some preliminary Overclocking testing with a promise of going more in depth later in a review to come in a week or two.

I thought the review was well done but for me personally I could have done without the 4k testing. All the 18 members gaming at 4k probably appreciated that.

On one hand I thought because of the downsampling feature it may have been helpful but I have a feeling it wont be a direct reflection of native 4k gaming since it's implementation was different.

Some reviews showed what I expected. Overclocking the Maxwell GPU's allows for some amazing gains.

Here is one review http://www.guru3d.com/articles_pages/gigabyte_geforce_gtx_980_g1_gaming_review,26.html

They got the Gigabyte G1 Gaming triple fan GTX 980 card to 1547 Boost!!

Here is where they got the GTX 970 version of the same card to 1516 Boost!!

http://www.guru3d.com/articles_pages/gigabyte_geforce_gtx_970_g1_gaming_review,26.html

The bad thing is that they are not Brent and I prefer Brent and the gangs Methodology of testing here more. What that type of OC does in some synthetic benchmarks is irreleveant to me.



I'm not here to criticize the review done here but just state my preference. Brent deserves our appreciation for busting his ass and bringing us a amazing review and losing sleep because of it.

I hope we all used the commissioned buying links also, I did with my pair of GTX 980s which are arriving tomorrow since they are shipping from newegg in Edison, NJ just a few towns away from me :D
 
Last edited:
On that note, anyone know of a review where they overclocked the cards and did several real world gaming comparisons and reported temps/fan noise and voltage?
 
I've only had the card since mid-day Monday this week, NDA lifted 7:30PM Pacific Thursday.

It took me all the time I had, plus more (didn't sleep the night before) to get what you see in the review online on time.

Impossible to do overclocking considering everything else that had to be done. I am not just going to throw some half arse overclocking job up just to have something. It wouldn't be accurate, or good, and could paint a wrong picture of overclocking. Give me the proper time to do a proper overclocking review and you will see a full overclocking evaluation not just overclocking 980 but also overclocking 780 Ti and 290X to maximum potential, and making comparisons. That is more informative to everyone.
 
Brent_Justice said:
I've only had the card since mid-day Monday this week, NDA lifted 7:30PM Pacific Thursday.

It took me all the time I had, plus more (didn't sleep the night before) to get what you see in the review online on time.

Impossible to do overclocking considering everything else that had to be done. I am not just going to throw some half arse overclocking job up just to have something. It wouldn't be accurate, or good, and could paint a wrong picture of overclocking. Give me the proper time to do a proper overclocking review and you will see a full overclocking evaluation not just overclocking 980 but also overclocking 780 Ti and 290X to maximum potential, and making comparisons. That is more informative to everyone.
Click to expand...

Brent I understand and thanks so much for the review, it was extremely well done. I guess Kyle or you decided on 4k testing. I would have preferred overclocking, or as mentioned above preliminary overclocking 1 or 2 games and maybe a loop for stability and a parragraph of your comments and a mention that you will go more in depth later in a week or two. because it's more important to me. 4k is such a small demographic on this site and others.

Please don't mistake that for me in any way saying your work didn't kick ass.
 
You must log in or register to reply here.
Back
Top