Sapphire to Launch RX 5500 XT Nitro + Special Edition with ARGB Fans - $259

[Totally]

LMAO. So this card costs 8% more then the RX 480 did at launch, and likely wont be more then 8% faster, given where the 5500 lands.

but I'm sure many people will say what a GREAT card this is, despite not moving the price/perf at ALL from 3 years ago, while lambasting Nvidia for the same thing. This thing should have been $199 at the most, given where it falls in the product stack, and even that is a hard sell. The 1660 super can be had for as low as $229, when the cheapest models are in stock, and reliably at $240-250, and even if the 5500xt is 20% faster then the 5500 TPU reviewed, the 1660 super will still be faster and a bit cheaper.

Good card, abysmal price point.

What are you on about? The RX 480 Nitro+ OC was $279 iirc, and $229 for the RX 480 Nitro non "+" <‐ I bought that one and OC'd the crap out of it before selling it for a nice profit during the craze.

Anyway, don't know how fast that card is going to be but's definitely going to be $20 cheaper(srp) than it's predecessor.

 

[Casecutter]

Well being this is Sapphire's "Top Self" with all the pretty LED spinning fans I guess $260 might be the "ask price" or is just a "place holder" to judge reaction, but yeah I couldn't see plunking down that price for a card that even if OC'd headily (closure to 2000 MHz) could basically get above a nicer GTX 1660. I cant see it in "Super" Territory.

That said, Navi 14 being 45% smaller than a TU116, and IDK... I think that there's not a ton of the "Super" that Nvidia has at least to this point to keep-up binning based on whatever they don't have to geld in the memory controller to let them work on GDDR6.

I think if AMD/RGT want's to go aggressive they'll work from a $200 MSRP, but I think it will be like $220 MSRP and many cards will be $230-240 at release, but can start working prices down. Remember the RX 590 started a $280, while today if you spend $200 for a nice one you didn't shop around.

 

[Zach_01]

5500 = 120~130$
5500XT = 170~180$
5600? = 220~230$

And nVidia can go home from the 100 to 450$ price point... but I somehow doubt that this will be the case/prices

 

[cucker tarlson]

I have a the sapphire 5700xt nitro+, and it has the same exact design, the cooler extends beyond the PCB, so I took the initiative, removed my 360mm AIO, put on an old thermalright macho air cooler I have then used tape to cover the end of the air cooler to simulate as if the PCB extend all the way to the end of the cooler, and then stress tested my system for a half hour. I did this with the tape on the GPU and the tape off. The frequency of the GPU and the CPU were kept constant as well as the RPMs of all my case fans and the results were that my CPU Temps were within 3 tenths of a degree Celsius for both tests... Your turn



Read my post above... I ACTUALLY tested your theory and there was no change in my CPU temperature

Lol,5700 nitro is a triple fan card,the open space is much smaller and not positioned right under the tower and its got a larger heatsink too.
I somehow doubt that you actually did any od this,you dont seem to understand.

 

[Sir Alex Ice]

AMD and especially Sapphire are complete idiots when pricing their new products. They make shitty VGA when it comes to performance vs. nvidia and make it more expensive.
No worries, they have a lot of margin behind their products, as evidenced by the BF deals where online shops in my country offered nearly 30% discount and I am sure they still had at least another 10% margin for themselves.

This policy is not good for the distributors, they have no reason to carry such a high risk to lose money on expensive VGA with crappy performance.

 

[dj-electric]

AMD and especially Sapphire are complete idiots when pricing their new products. They make shitty VGA when it comes to performance vs. nvidia and make it more expensive.
No worries, they have a lot of margin behind their products, as evidenced by the BF deals where online shops in my country offered nearly 30% discount and I am sure they still had at least another 10% margin for themselves.

This policy is not good for the distributors, they have no reason to carry such a high risk to lose money on expensive VGA with crappy performance.

Here's why i don't have an issue with Sapphire - They can create exotic and over priced versions of a card, but at the same time they will also compete for being the best cheapest option like Pulse edition. on RX 5700 series it does it wonderfully and don't expect anything else on RX 5500 XT.

 

[tomkocur]

I could draw you a schematic but is this so hard to grasp? In any typical case you have airflow going from front to back, this card places a new flow at a 90 degree angle to that, while any normal open cooler does that indirectly, barely impeding the case airflow, basically just leaking the hot air into it.

This is a GPU fan, with a strong directional airflow, too. Def gonna cost you a few C for no reason, well one: a higher price.

Actually, no. Airflow goes from front bottom/bottom to top rear/top in standard cases - with the down -> up direction being the most efficient, as warm air has a tendency to go upwards. Outtake fans are either behind or above the CPU tower cooler, so an outgoing air ALWAYS goes through the cpu cooler. So the only difference between a cooler that's blasting air towards the GPUs PCB and a cooler that can blast air freely upwards, is that in the latter case GPU cooling will be more efficient with minimal difference to the CPU temperature.
It's mostly stock coolers with radial fans that direct the outgoing airflow to the back of the case.
Also don't forget there's a thing that's called thermal capacity. If a GPU wakes up from 0dB and starts blasting hot air at the cpu cooler, it wouldn't cause a dramatic "sudden" rise of the CPU temp. Air is not that effective and heating up 500g+ of aluminimum and copper takes some time.

 

[Xaled]

Show an initial high price then announce a better price at launch? Just like what they did with 57xx's?

 

[TheinsanegamerN]

What are you on about? The RX 480 Nitro+ OC was $279 iirc, and $229 for the RX 480 Nitro non "+" <‐ I bought that one and OC'd the crap out of it before selling it for a nice profit during the craze.

Anyway, don't know how fast that card is going to be but's definitely going to be $20 cheaper(srp) than it's predecessor.

What is so confusing to you? This card, at $259, is an ABYSMAL value, not moving the $/perf meter at all from 3 years ago. That $279 480 was also an abysmal value/$ card.

 

[Vayra86]

Actually, no. Airflow goes from front bottom/bottom to top rear/top in standard cases - with the down -> up direction being the most efficient, as warm air has a tendency to go upwards. Outtake fans are either behind or above the CPU tower cooler, so an outgoing air ALWAYS goes through the cpu cooler. So the only difference between a cooler that's blasting air towards the GPUs PCB and a cooler that can blast air freely upwards, is that in the latter case GPU cooling will be more efficient with minimal difference to the CPU temperature.
It's mostly stock coolers with radial fans that direct the outgoing airflow to the back of the case.
Also don't forget there's a thing that's called thermal capacity. If a GPU wakes up from 0dB and starts blasting hot air at the cpu cooler, it wouldn't cause a dramatic "sudden" rise of the CPU temp. Air is not that effective and heating up 500g+ of aluminimum and copper takes some time.

Good points, sir. And yes I'm sure it won't be horrible; but the million dollar question here rather is 'why'. It doesn't look mighty effective.

But more importantly, the card could have been smaller.

I have a the sapphire 5700xt nitro+, and it has the same exact design, the cooler extends beyond the PCB, so I took the initiative, removed my 360mm AIO, put on an old thermalright macho air cooler I have then used tape to cover the end of the air cooler to simulate as if the PCB extend all the way to the end of the cooler, and then stress tested my system for a half hour. I did this with the tape on the GPU and the tape off. The frequency of the GPU and the CPU were kept constant as well as the RPMs of all my case fans and the results were that my CPU Temps were within 3 tenths of a degree Celsius for both tests... Your turn



Read my post above... I ACTUALLY tested your theory and there was no change in my CPU temperature

You did all that work, but forgot to take a picture? Convince me you're not lying. I don't buy strong statements like that in plain text.

 

[INSTG8R]

Great so you pay a hefty premium for a backplate that doesn't even cover the whole heatsink

Why is the card even that long?!

It worked great for the Tri-X Fury ¯\_(ツ)_/¯ when I switched to my “normal“ Vega I had to rethink my case cooling/fan placement. Basically you’ve got a 1/4 of the fin stack unimpeded to dissipate heat.

 

[Ruru]

Prices are insane these days. 259 USD for a mid-end card?! Back in the day you could get a high-end card like 8800 GTS or HD 4870 with that price.

 

[djisas]

Prices are insane these days. 259 USD for a mid-end card?! Back in the day you could get a high-end card like 8800 GTS or HD 4870 with that price.

For that price, i bought cards like 8800GTS 512, HD4850 and HD6850, all great cards...
200$ for entry level performance is pretty bad these days...

 

[Ruru]

For that price, i bought cards like 8800GTS 512, HD4850 and HD6850, all great cards...
200$ for entry level performance is pretty bad these days...

Exactly, that's what I mean. And those higher tier mid-end cards also brought better performance back for the money in the day, let's say GTX 460, GTX 660, HD 7870 for example when comparing to the price difference between those series' high end models.