Do Tom's Hardware actually review all the cards in their graphs.

[Kaapstad]

I have just been looking at a review for a Gigabyte Aorus Xtreme 2080 Ti card on Tom's Hardware and noticed the graphs contained performance figures for the RTX Titan and Titan V. Some of the figures looked a bit off so I tried to find their reviews for the Titans but was unable to.

Maybe it is me going blind or failing that where are Tom's Hardware getting their figures from for these cards?

https://www.tomshardware.com/reviews/gigabyte-aorus-geforce-rtx-2080-ti-xtreme-11g,5953-3.html

 

[kitfit1]

I think you've answered your own question Kaapstad. You couldn't find the reviews on Tom's Hardware, because of course they didn't review the cards in the first place. More than likely just pulled the figures from another site.

 

[LePhuronn]

Tom's didn't legitimately review Nvidia cards? Wow. If there were any results pasted in from other sources I'd fully expect them to be AMD flavoured.

 

[subbytna]

Ask them. #threadclosed

 

[Silent_Scone]

I have just been looking at a review for a Gigabyte Aorus Xtreme 2080 Ti card on Tom's Hardware and noticed the graphs contained performance figures for the RTX Titan and Titan V. Some of the figures looked a bit off so I tried to find their reviews for the Titans but was unable to.

Maybe it is me going blind or failing that where are Tom's Hardware getting their figures from for these cards?

https://www.tomshardware.com/reviews/gigabyte-aorus-geforce-rtx-2080-ti-xtreme-11g,5953-3.html

Might want to check Anandtech, or a number of other sites, too. They are part of the same conglomerate (Purch Group).

 

[Kaapstad]

Ask them. #threadclosed

I think what I am asking in the OP is a fair point as people sometimes get misled by dodgy tech sites.

This thread has far more relevance than some of the NVidia bashing ones I have seen lately.

 

[Maldoror]

It is an interesting question and Tom's have been doing some odd things lately (most memorably the 'How much of your life can you risk not being ray traced before you die' and all that Just Buy It stuff).

They do state for this review:

"For Nvidia, that means testing the two new games with build 417.22 (all of the Gigabyte card’s numbers are generated with that driver, too). The Founders Edition cards are tested with 416.33 (2070) and 411.51 (2080 and 2080 Ti). Older Pascal-based boards are tested with build 398.82. Titan V’s results were spot-checked with 411.51 to ensure performance didn’t change. AMD’s cards utilize Crimson Adrenalin Edition 18.8.1 (except for the Battlefield V and Wolfenstein tests, which are tested with Adrenalin Edition 18.11.2)."

So it sounds like they're using previous results with the Titan V and then 'spot checked' to verify the general performance. But technically you could argue - given that they have no review - that the careful phrasing above allows for the Titan V's results to be simply extrapolated from data/specs and that this extrapolated performance difference was what was 'spot-checked'. No mention of their Titan RTX numbers either.

At any rate, I do find it bizarre that they benchmark cards across so many games and yet never publish a review. Surely after doing all the hard work, even a simple review/opinion with some blurb would be worth it for them to generate hits?

 

[Silent_Scone]

It is an interesting question and Tom's have been doing some odd things lately (most memorably the 'How much of your life can you risk not being ray traced before you die' and all that Just Buy It stuff).

They do state for this review:

"For Nvidia, that means testing the two new games with build 417.22 (all of the Gigabyte card’s numbers are generated with that driver, too). The Founders Edition cards are tested with 416.33 (2070) and 411.51 (2080 and 2080 Ti). Older Pascal-based boards are tested with build 398.82. Titan V’s results were spot-checked with 411.51 to ensure performance didn’t change. AMD’s cards utilize Crimson Adrenalin Edition 18.8.1 (except for the Battlefield V and Wolfenstein tests, which are tested with Adrenalin Edition 18.11.2)."

So it sounds like they're using previous results with the Titan V and then 'spot checked' to verify the general performance. But technically you could argue - given that they have no review - that the careful phrasing above allows for the Titan V's results to be simply extrapolated from data/specs and that this extrapolated performance difference was what was 'spot-checked'. No mention of their Titan RTX numbers either.

At any rate, I do find it bizarre that they benchmark cards across so many games and yet never publish a review. Surely after doing all the hard work, even a simple review/opinion with some blurb would be worth it for them to generate hits?

I think what I am asking in the OP is a fair point as people sometimes get misled by dodgy tech sites.

This thread has far more relevance than some of the NVidia bashing ones I have seen lately.

Was merely pointing out the fact, before the thread turned sour. Purch owns a number of review sites, so for the sake of comparison, there doesn't necessarily need to be a review on Tomshardware for every card. There will be enough data across the board to work with.

 

[Kaapstad]

Was merely pointing out the fact, before the thread turned sour. Purch owns a number of review sites, so for the sake of comparison, there doesn't necessarily need to be a review on Tomshardware for every card. There will be enough data across the board to work with.

Even if that is what they are doing it is obviously not working that well.

As someone who has 2080 Ti, RTX Titan and Titan V cards all in the same place I can see the results in some of their graphs are wrong.

I think the proper way to do a review is to use the same PC, latest drivers and same conditions.

 

[NMBlack]

Even if that is what they are doing it is obviously not working that well.

As someone who has 2080 Ti, RTX Titan and Titan V cards all in the same place I can see the results in some of their graphs are wrong.

I think the proper way to do a review is to use the same PC, latest drivers and same conditions.

Is every article on TH generated in-house, or do they take freelance submissions? If it's the latter, maybe some of the GPU aficionados here could contribute review articles?

 

[Nasher]

I don't think Tom Hardware is what it once was. The suspicion is they are on the take.

 

[WantoN]

I stopped respecting Tom’s Hardware a long time ago, their ‘Just Buy It’ article was just the final nail in the coffin.

 

[Maldoror]

Even if that is what they are doing it is obviously not working that well.

As someone who has 2080 Ti, RTX Titan and Titan V cards all in the same place I can see the results in some of their graphs are wrong.

I think the proper way to do a review is to use the same PC, latest drivers and same conditions.

Indeed. Even taking their testing description at face value, we're still talking about driver versions that are months apart, games that will have had engine updates and patches during that time and generally an odd testing methodology.