AMD 7600 reviews

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
28,843
21,643
146
This is the exact mentality nVidia and AMD want the consumer to have. And if the consumer continues to purchase GPUs with this lack of generational improvement and rising cost they will continue to milk every last penny from the consumer's backside. At some point the consumer has to take a stand.
They are treating us like the proverbial frogs in the pot.

Unnecessary content removed.
 
Last edited:

Heartbreaker

Diamond Member
Apr 3, 2006
4,263
5,260
136
Indirectly, to me it feel they put all the R&D preparing the chiplets and left most of the rest untouched.
So next round, maybe?

Core units were reworked for much more FP capability, so it's not like the only R&D was on chiplets.

You can't expect the basic CU to have decent IPC gains every generation. Where would they come from?

The biggest issues with monolithic N33 is that it has the same CU count, and the same process, leading to about the same performance.

While N31 and N32 with chiplets, both have more CUs.

It's almost like the chiplets are doing their job and enabling more die devoted to core logic, so you get more CU's.

I think the Memory/Cache chiplets are the best thing about this generation.
 

biostud

Lifer
Feb 27, 2003
18,407
4,968
136
RDNA3 reminds me of Bulldozer's "dual" cores that didn't really work either.
 

jpiniero

Lifer
Oct 1, 2010
14,845
5,457
136
That is my point: N33 and N32 should both have been monoliths. At least a monolith N32 would have the advantage of a lower BOM . As Navi33 is a about 15% smaller than Navi23, so a monolith N32 on 6nm might have been around 290mm² - a size where yields should still have been good - ergo chiplets don't really save anything especially once packaging costs and the higher power overhead are take into account.

Cache/IO scaling blows on N5, and N3E has no scaling at all. I wouldn't be surprised if the MCD are reused for RDNA4.
 
Reactions: Tlh97 and KompuKare

Dribble

Platinum Member
Aug 9, 2005
2,076
611
136
This is the exact mentality nVidia and AMD want the consumer to have. And if the consumer continues to purchase GPUs with this lack of generational improvement and rising cost they will continue to milk every last penny from the consumer's backside. At some point the consumer has to take a stand.
What are they going to do - make a new gpu company and start pumping out gpu's at a lower margin for the common good? Realistically neither "the consumer" nor people on this forum will do anything. Pricing is what it is.
 

Panino Manino

Senior member
Jan 28, 2017
847
1,061
136
Aside from AMD humiliating themselves again with a last minute price change, I feel that the reviews are a bit too harsh?
Reviewers forgot that this is the 7600 non-xt? I got the impression they were expecting the performance from the 7700 line.

The price could still be lower but I think they have to accept that those old prices will never come back.
 

Saylick

Diamond Member
Sep 10, 2012
3,391
7,156
136
Cache/IO scaling blows on N5, and N3E has no scaling at all. I wouldn't be surprised if the MCD are reused for RDNA4.
The MCDs don't appear to be the culprit for RDNA3's lack of performance anyways, so I agree that AMD should focus their efforts on the GCD.
 

Mopetar

Diamond Member
Jan 31, 2011
8,024
6,479
136
RDNA3 reminds me of Bulldozer's "dual" cores that didn't really work either.

I think AMD doesn't have their drivers in place for the feature yet. NVidia moved to this same model with Ampere (at least for FP since they already code do mixed FP/int) where they "doubled" the cores in the same way.

Obviously it doesn't double performance outside of some pure compute workloads, but there's probably at least a 10% boost for AMD to extract. My guess is that they only added it when they found out NVidia was and realized it's probably a good idea, but that leaves them playing catch-up as NVidia obviously would have spent a lot more time thinking about such a feature and how to leverage it.

Best-case scenario is that there's nothing broken in the hardware and that we get a bit of fine wine down the road when they get it all figured out well enough for the gains to show up in RDNA4.
 

Hans Gruber

Platinum Member
Dec 23, 2006
2,223
1,156
136
Aside from AMD humiliating themselves again with a last minute price change, I feel that the reviews are a bit too harsh?
Reviewers forgot that this is the 7600 non-xt? I got the impression they were expecting the performance from the 7700 line.

The price could still be lower but I think they have to accept that those old prices will never come back.
If you are not first, you're last. AMD has always been a good clip behind Nvidia in the GPU segment. The last time they offered a compelling value was the RDNA 1 5700 and 5700xt based on price and performance. Which is probably why people still talk about the performance of the 5700xt on this forum regularly.

Then there is the AMD driver issue. They fix their drivers and optimize them. You just have to wait a year or two for the performance gains.

There is no longer any significant uplift in performance for TSMC node shrinks. The only benefit is power efficiency on newer nodes.
 

Joe NYC

Platinum Member
Jun 26, 2021
2,333
2,947
106
My 6600XT maxes out at 130W. This card is the 💩 I expected. Good to see reviewers suddenly "discovering" the 6700 still exist, at least here in the U.S. and cost the same money. That's why I said this card was pointless when I saw the specs. It is AMD, so there might be a little fine wine in the tank, but nothing is going to save this turd.

It has about 15% smaller die than Navi 23, N6 vs. N7 to offer additional fabrication savings, and performance increase between 5% and 10%.

People saying there is no point in introducing this card are not thinking straight.

They are comparing last gen card that have been partially written off and therefore heavily discounted vs. what is sustainable and manufacturable going forward.
 
Reactions: Tlh97 and DooKey

blckgrffn

Diamond Member
May 1, 2003
9,198
3,185
136
www.teamjuchems.com
It has about 15% smaller die than Navi 23, N6 vs. N7 to offer additional fabrication savings, and performance increase between 5% and 10%.

People saying there is no point in introducing this card are not thinking straight.

They are comparing last gen card that have been partially written off and therefore heavily discounted vs. what is sustainable and manufacturable going forward.
I am saying they could have just done a real 6600 die (no cuts) on N6 months and months ago, poured that 5-10% performance uplift into the same 100W envelope and have been done with it. It would be so close to the same ball park as to be difficult to distinguish in actual games, be nearly the same die size and importantly would have been that much more profitable at $200 MSRP all the while. The RNDA2 drivers are solid and they could have done a model number bump of some sort as well. It wouldn't have changed the fact that its the best card @ $200 either, because nvidia appears very disinterested.

The 7600 doesn't even replace this card well because it uses 50% more power, which is non-trivial. It's got a 6600xt/6650xt power budget which is a bit silly because those were already powered a bit past nominal.
 

Joe NYC

Platinum Member
Jun 26, 2021
2,333
2,947
106
I am saying they could have just done a real 6600 die (no cuts) on N6 months and months ago, poured that 5-10% performance uplift into the same 100W envelope and have been done with it. It would be so close to the same ball park as to be difficult to distinguish in actual games, be nearly the same die size and importantly would have been that much more profitable at $200 MSRP all the while. The RNDA2 drivers are solid and they could have done a model number bump of some sort as well. It wouldn't have changed the fact that its the best card @ $200 either, because nvidia appears very disinterested.

The 7600 doesn't even replace this card well because it uses 50% more power, which is non-trivial. It's got a 6600xt/6650xt power budget which is a bit silly because those were already powered a bit past nominal.

Agreed, maybe AMD should have gone with N6 with their 6*50 refresh cards, and spent extra time perfecting RDNA3. But that is already water under the bridge, we can't change the past.

I have to read some more reviews to get more info on the power consumption. I also have a 6600 XT card, and it is very power efficient. So it would be disappointing to go backward from there.
 

jpiniero

Lifer
Oct 1, 2010
14,845
5,457
136
The 7600 doesn't even replace this card well because it uses 50% more power, which is non-trivial. It's got a 6600xt/6650xt power budget which is a bit silly because those were already powered a bit past nominal.


Efficiency is pretty good, esp with only being on N6.
 

biostud

Lifer
Feb 27, 2003
18,407
4,968
136
I think AMD doesn't have their drivers in place for the feature yet. NVidia moved to this same model with Ampere (at least for FP since they already code do mixed FP/int) where they "doubled" the cores in the same way.

Obviously it doesn't double performance outside of some pure compute workloads, but there's probably at least a 10% boost for AMD to extract. My guess is that they only added it when they found out NVidia was and realized it's probably a good idea, but that leaves them playing catch-up as NVidia obviously would have spent a lot more time thinking about such a feature and how to leverage it.

Best-case scenario is that there's nothing broken in the hardware and that we get a bit of fine wine down the road when they get it all figured out well enough for the gains to show up in RDNA4.
I’m all for hoping, but since it is the core of RDNA3, it seems weird that their drivers aren’t able to use it at this point. :/
 
Reactions: Joe NYC

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
28,843
21,643
146
It has about 15% smaller die than Navi 23, N6 vs. N7 to offer additional fabrication savings, and performance increase between 5% and 10%.

People saying there is no point in introducing this card are not thinking straight.

They are comparing last gen card that have been partially written off and therefore heavily discounted vs. what is sustainable and manufacturable going forward.
So introduce it when most of the 6600 series stock is gone for a lower price. While there seem to be few models of the 6700 left; a card with more vram for the same money. The 6600 series is still chock-a-block. With 6600XTs and 6650XTs often going for sub $225, and the 6600 $200 sometimes $189, the 7600 value isn't there. That might not be the best deal for AMD and partners, but it is for us.

With the 67xxXT series dropping to $320 it is even more pressure. If you can't stretch the budget, 6600 is the best value. If the prices of SSDs and DDR dropping as they have created new wiggle room in the budget, the 6700 series is the winner. They could not have picked a worse time to launch with a MSRP over $225 IMO. It's in no man's land.

By releasing it now with bad pricing, they generated more negative brand image, and left a bad taste in gamers mouths. If AMD isn't putting their foot in their mouth, they are shooting themselves in said foot.
 

blckgrffn

Diamond Member
May 1, 2003
9,198
3,185
136
www.teamjuchems.com

Efficiency is pretty good, esp with only being on N6.
Meh, I don't agree with Tom's (Jared's) conclusions generally, but I like their data generally.



It's not a great look, and for all its foibles the 4060ti is both faster and more efficient and the 4060 will likely be at least as fast and significantly more efficient. In a world where Ampere and RDNA2 don't exist, the 4060 vs 7600 is likely a nice and tidy win for the green team.

Which is a loss for all of us who want more perf in the $300-$500 segment. Which seems like a really reasonable, cost of a console type segment.
 
Reactions: Tlh97 and Rigg

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
28,843
21,643
146
Do you know if Newegg has a 30-day price policy? I bought that one 5/6, for $379.

John
For all the horror stories about da egg now, I have had great experiences when I needed to contact CS.

There is a 30 day hassle free return policy. You may be able to get the rep to clear with their super, to let you have the credit instead of having to return and reorder the card. I can't imagine they want to eat the open box return plus shipping you can put them through.
 

fralexandr

Platinum Member
Apr 26, 2007
2,249
201
106
www.flickr.com
Do you know if Newegg has a 30-day price policy? I bought that one 5/6, for $379.

John
Looks like Newegg has a 30 day return policy. I asked a customer rep about a price match a while back on a nvme drive and from what I remember they gave me a promo credit that I think had an expiration date in something like 90 days? which I was ok with. It was a bit over a year ago, so the details are a bit fuzzy.
 

DAPUNISHER

Super Moderator CPU Forum Mod and Elite Member
Super Moderator
Aug 22, 2001
28,843
21,643
146
Meh, I don't agree with Tom's (Jared's) conclusions generally, but I like their data generally.
Tom's has botched a few reviews lately where they had to update them. Which made them even harder to read. I put little stock in their testing at this point.
 

Panino Manino

Senior member
Jan 28, 2017
847
1,061
136
I think AMD doesn't have their drivers in place for the feature yet. NVidia moved to this same model with Ampere (at least for FP since they already code do mixed FP/int) where they "doubled" the cores in the same way.

Obviously it doesn't double performance outside of some pure compute workloads, but there's probably at least a 10% boost for AMD to extract. My guess is that they only added it when they found out NVidia was and realized it's probably a good idea, but that leaves them playing catch-up as NVidia obviously would have spent a lot more time thinking about such a feature and how to leverage it.

Best-case scenario is that there's nothing broken in the hardware and that we get a bit of fine wine down the road when they get it all figured out well enough for the gains to show up in RDNA4.

Even if the engineers explain to me I'll never understand this "stupid" decision. Why put so much time and resources engineering a feature that they will not be able to make use? Will be left there just wasting space? And specially this feature, after they had made so much progress solving this exactly problem with the changes from RDNA1.
RDNA3 would be so much better (IMO) if they had kept the CUs "simpler" and just put more of them in the dies.

There's so much madness into how AMD makes decisions.
I really hoped that the graphics side that finally get on track like the CPU side that it's making steady and consistent progress with the Zen architecture.
How probably it is that RDNA4 will finish derail it?

Anyway, the price situation is what it is.
It's easy to complain and demand AMD to ask lower prices but this is unviable, they need to make money.
 

Joe NYC

Platinum Member
Jun 26, 2021
2,333
2,947
106
So introduce it when most of the 6600 series stock is gone for a lower price. While there seem to be few models of the 6700 left; a card with more vram for the same money. The 6600 series is still chock-a-block. With 6600XTs and 6650XTs often going for sub $225, and the 6600 $200 sometimes $189, the 7600 value isn't there. That might not be the best deal for AMD and partners, but it is for us.

With the 67xxXT series dropping to $320 it is even more pressure. If you can't stretch the budget, 6600 is the best value. If the prices of SSDs and DDR dropping as they have created new wiggle room in the budget, the 6700 series is the winner. They could not have picked a worse time to launch with a MSRP over $225 IMO. It's in no man's land.

By releasing it now with bad pricing, they generated more negative brand image, and left a bad taste in gamers mouths. If AMD isn't putting their foot in their mouth, they are shooting themselves in said foot.
AMD needed N33 for the latest generation of laptops, so they had to be put into production. And once in production and sufficient stock available, there is no point in holding back the desktop release.

Pricing is not terrible:
- N23 $379
- N33 $269, likely at more modest margins than N23. Not a lot of price premium there.

Pricing seems quite realistic from AMD / AIB cost POV. You may question performance and want more of it, but not necessarily price, given the BOM.

One thing we can be sure of, is that AMD is not going to pricing of next 1 million cards based on a price of a random retailer with stock of 3 units of the last gen product.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |