Talk:Radeon HD 5000 series

Computing Low‑importance

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing
Low	This article has been rated as Low-importance on the project's importance scale.
	This article is supported by Computer hardware task force.

Rumours

Based on 40 nm manufacturing process by TSMC, the RV870 will see memory amount to the magnitudes of GBytes, and possible additions of texture units, texture caches and ROPs. --202.40.157.145 (talk) 03:46, 5 May 2008 (UTC)[reply]

initial planning 1000 SPUs, continue to be based on the Radeon R600 architecture as the fourth generation of ATI's Unified Shader architecture (First generation being Xenos/C1, second generation being Radeon R600 and RV670 families, the third generation being Radeon R700 family). --202.40.157.145 (talk) 05:28, 21 May 2008 (UTC)[reply]

Dual GPU version also available, with memory sharing across two GPUs.

¹ Unified Shaders (Vertex shader/Geometry shader/Pixel shader) : Texture mapping unit : Render Output unit

Model	Year	Code name	Fab (nm)	Bus interface	Memory max (MiB)	Reference clock rate		Config core¹	Fillrate		Memory			Graphics library support (version)		Notes
Model	Year	Code name	Fab (nm)	Bus interface	Memory max (MiB)	Core (MHz)	Memory (MHz)	Config core¹	Texture (GT/s)	Pixel (GP/s)	Bandwidth (GB/s)	Bus type	Bus width (bit)	DirectX	OpenGL	Notes
Cypress	Q3 2009	Cypress	40	PCIe 2.0 x16	1024	1000	1250	2000(400x5)96:48	96	48	160	GDDR5	256	11
Unknown	Q4 2009		40	PCIe 2.0 x16	2048	900	1200	2x [2000(400x5):96:48]	172.8	86.4	307.2	GDDR5	256	11		Dual GPU solution, memory sharing

is this table confirmed ..i heard that RV870 (NOT R800) Continued 1200~1600 Stream Processor Max ..that table said 2000 SPs .Salem F (talk) 16:07, 31 August 2009 (UTC)[reply]

Merge with article Radeon HD5870

There is an article with the name Radeon HD 5870. Its contents are similar. I think these two articles should be merged. Cristan (talk) 08:51, 25 September 2009 (UTC)[reply]

surface area

it would be ideal to have the surface area digits for each chip —Preceding unsigned comment added by Em27 (talk • contribs) 21:51, 24 November 2009 (UTC)[reply]

Lemma

I'm wondering why the lemma is Evergreen (GPU family) while it is the code number of the chip family for all previous generations. Radeon R100 - Radeon R700 --Mewtu (talk) 23:51, 2 December 2009 (UTC)[reply]

Mobile Section?

Can we get a section of the mobile graphic cards in the 5000 series?76.21.122.234 (talk) 01:49, 8 May 2010 (UTC)[reply]

There is no need for a simplified chipset table in the article.

Unnecessary content forking at best, where there are already links (in the chipset table section) to direct people who are interested into the subject to read the relevant seection in the GPU list, so that subsection was removed. --121.202.93.67 (talk) 20:29, 21 June 2010 (UTC)[reply]

Highest number first

Why are the cards in a reverse order? --82.215.247.45 (talk) 21:22, 4 July 2010 (UTC)[reply]

Lemma impractical - should be "ATI HD5000 series" or sth. alike

The lemma is a catastrophe ! Only hard-core computer geeks know the internal codenames of GPU families, so how is the average user supposed to find this wiki page ? The German wikipedia, for comparison, has this GPU-family listed under its sales name, namely "ATI Radeon HD 5000 series". This way any user can find it quickly and easily: http://de.wikipedia.org/wiki/ATI-Radeon-HD-5000-Serie -- Alexey Topol (talk) 11:59, 3 November 2010 (UTC)[reply]

My Edits are Valid

ATi is ATi, AMD is AMD.

Sure, AMD acquired ATi in 2006, BUT AMD does NOT make the GPUs. ATi does. AMD is a CPU company, not a GPU one, and they merely Slapped their Logo on recent GPUs: Nothing more.

I will proceed to correct these pages. — Preceding unsigned comment added by 174.117.204.241 (talk) 12:19, 1 June 2011 (UTC)[reply]

Correct the entries please, it is about time...

In the article about the HD 5000 Series, section Architecture, there is the following information: This article is about all products under the Radeon HD 5000 Series brand. TeraScale 2 was introduced with this.

   A GPU implementing TeraScale 2 is found on Radeon HD 5830 and above branded products. [...]
   A GPU implementing TeraScale 1 is found on Radeon HD 5770 and below branded products. [...]

A TS2 GPU, HD 5830 and above means the Cypress chip and Hemlock (2x Cypress), a TS1 GPU, HD 5770 and below means the Juniper chip - what about Cedar and Redwood chips?? It seems this information differs from that about the GPU architecture, see the lemma TeraScale, TeraScale 2 section:

   TeraScale 2 (VLIW5) was introduced with the Radeon HD 5000 Series in the "Evergreen" series of GPUs. [...] 
   Evergreen chips: Cedar RV810, Redwood RV830, Juniper RV840, Cypress RV870 [...]

The question is: are Juniper, Cedar and Redwood Terascale 2 chips or not?? If so, all HD 5000 series video cards had TS2 implemented, not just a few (Cypress, Hemlock). This needs to be corrected by someone with knowledge about the matter. The german wp is listing all of them as TS2 chips by the way.

Also include a row for the Architecture into the products table provided, between Codename and Fab (nm), as it is found with the graphics cards table for the HD 6000 series GPUs. Sorting the architetures accordingly (if there are any differences) does help to get a quick overview about what technology was used in which products. — Preceding unsigned comment added by 88.151.79.149 (talk) 15:32, 15 September 2017 (UTC)[reply]

I don't think he should use a doubling based on a core scale design

Determine if he is using TeraScale 1 or TeraScale 2

They all have architectural improvements and support for DirectX 11/DirectCompute 11/Open cl in the TeraScale 2 architecture 61.216.108.177 (talk) 05:59, 21 September 2022 (UTC)[reply]

They both have TeraScale 2 architectural improvements (except double the core design scale/double-precision compute), and both have DirectX 11/DirectCompute 11/Opencl support starting in the TeraScale 2 architecture 61.216.108.177 (talk) 06:03, 21 September 2022 (UTC)[reply]

Is the 5770 really TeraScale 1 ?

This explicitly mentions the 5770 as TeraScale 2 https://wiki.riteme.site/wiki/TeraScale_(microarchitecture)#TeraScale_2

This doesn't mention the 5000 series https://wiki.riteme.site/wiki/TeraScale_(microarchitecture)#TeraScale_1

Which article is right?

--Tuxayo (talk) 23:37, 30 May 2019 (UTC)[reply]

They all have architectural improvements and support for DirectX 11/DirectCompute 11/Opencl in the TeraScale 2 architecture 61.216.108.177 (talk) 06:00, 21 September 2022 (UTC)[reply]

They both have TeraScale 2 architectural improvements (except double the core design scale/double-precision compute), and both have DirectX 11/DirectCompute 11/Opencl support starting in the TeraScale 2 architecture 61.216.108.177 (talk) 06:04, 21 September 2022 (UTC)[reply]