BRIAN K. SMITH, APPELLANT,

v.

STATE OF FLORIDA, APPELLEE

CASANUEVA, Chief Judge.

Brian K. Smith appeals the summary denial of his motion filed pursuant to Flori*1278da Rule of Criminal Procedure 3.850. We reverse and remand for further proceedings.

In his motion, Mr. Smith alleged that the method of comparative bullet lead analysis (CBLA), which had been used by an FBI analyst who testified at his murder trials in 1989 and 1990, was subsequently discredited by scientific research and abandoned by the FBI as unreliable. Mr. Smith cited the November 2007 joint-investigation by The Washington Post and CBS News’ “60 Minutes” describing research by William A. Tobin, former chief metallurgist for the FBI, and by the National Research Council of the National Academy of Sciences, which concluded that CBLA could not reliably support testimony that a bullet came from a particular box of ammunition. Mr. Smith alleged that at his trials, FBI Special Agent Donald Have-kost had offered such expert testimony as the only direct evidence linking him to the crimes where other evidence was weak and circumstantial. The Post-“60 Minutes” story reported that the FBI had discontinued use of CBLA in 2005 but did little to alert prosecutors and defense attorneys. Mr. Smith claimed that information in the Post-“60 Minutes” investigation, which was not available at the time of trial, constituted newly discovered evidence justifying consideration of his otherwise untimely rule 3.850 motion.

Newly discovered evidence is evidence that was unknown at the time of trial, could not have been discovered by the use of due diligence, and would probably produce an acquittal on retrial. Jones v. State, 709 So.2d 512, 521 (Fla.1998). Summarily denying Mr. Smith’s motion, the postconviction court did not have the benefit of our decision in Clark v. State, 995 So.2d 1112 (Fla. 2d DCA 2008). In Clark, the movant “asserted that newly discovered evidence show[ed] that a scientific theory advanced by the State at his trial [had] recently been discredited or abandoned.” Id. at 1113. This court held that the postconviction movant presented facially sufficient allegations that the new evidence was unknown at the time of trial and could not have been discovered by the use of due diligence. Id. Because his allegation that the evidence would probably produce an acquittal on retrial was not conclusively refuted by the record, we reversed the summary denial of the claim and remanded for further proceedings. Id. at 1113-14.

Clark controls here. Mr. Smith sufficiently alleged that evidence that CBLA has been discredited and abandoned was unknown at the time of his trials and could not have been discovered by the use of due diligence. The postcon-viction court did not reach Mr. Smith’s allegation that the evidence would probably produce an acquittal on retrial, nor did it consider any other issue related to Mr. Smith’s motion. We express no opinion on issues not addressed by the postconviction court and reverse and remand for further proceedings.1

Reversed and remanded.

*1279DAVIS, J., Concurs.

ALTENBERND, J., Concurs specially.

ALTENBERND, Judge, Specially

concurring.

I agree that this case is controlled by Clark v. State, 995 So.2d 1112 (Fla. 2d DCA 2008). I wrote a separate concurrence in Clark. Id. at 1114. I am still concerned that a change in scientific opinion within a relevant scientific community — a change that occurred almost twenty years after the trial in this case — may require a different postconviction analysis than the typical claim of newly discovered evidence.

In a typical postconviction claim of newly discovered evidence, the defendant alleges that certain evidence existed at the time of trial that was unknown to the defendant and his or her counsel and that it could not have been discovered with due diligence. See, e.g., Padron v. State, 769 So.2d 432, 433 (Fla. 2d DCA 2000). For example, a witness who was driving by the scene of a convenience store robbery and did not stop to identify herself at the time of trial may be a source of newly discovered evidence when defense counsel inadvertently discovers the witness several years later. See, e.g., Clugston v. State, 765 So.2d 816 (Fla. 4th DCA 2000) (setting aside the defendant’s conviction based on newly discovered evidence that an unknown eyewitness came forward years after the conviction and identified a different perpetrator as the shooter).

This case involves evidence that did not exist at the time of trial. From the file, it appears that the evidence admitted at trial was properly admitted scientific evidence and scientific expert opinion that passed muster under a Frye2 analysis at that time. Mr. Smith is not arguing that the FBI analyst intentionally misrepresented the facts or his opinion at the time of trial. Cf. Commonwealth v. Kretehmar, 971 A.2d 1249 (Pa.Super.Ct.2009) (holding use of CBLA evidence did not warrant postcon-viction relief where the evidence was not a misrepresentation and did not deny defendant fair and impartial trial).

Since the time of the trial, the scientific community, including scientists working for the FBI, has changed its assessment of the methods or theories underlying comparative bullet lead analysis. Although such evidence was apparently used in as many as 2500 trials throughout the United States between 1963 and 2004 according to the materials provided to us by Mr. Smith,3 the relevant scientific community no longer believes such evidence is reliable.

The new discovery is not exactly new evidence;4 it is a new circumstance under which evidence that was legally admissible *1280at the time of trial would no longer be legally admissible today. Thus, at any new trial, there would not be a body of newly discovered evidence to present that favored Mr. Smith. Instead, it would be the same trial he had in 1989 without some or all of the testimony of the FBI analyst.5

I agree that there comes a point in a case like Mr. Smith’s where the legitimacy of even a twenty-year-old criminal conviction cannot withstand a major shift in scientific thought.6 I am hesitant, however, to reopen cases twenty years after a conviction merely because evidence that was properly admitted would not be admitted under current evidentiary standards. It is very hard for the State to retry a case and prove that case beyond a reasonable doubt when evidence is stale by a full generation.

Intuitively, it seems to me that when a defendant is arguing that evidence properly admitted at his trial would no longer be admissible at trial due to advances in scientific knowledge or theory, the defendant should have a heavy burden to establish that the admitted evidence was critical to the conviction. This seems particularly true in the case of comparative bullet lead analysis. The shift in scientific thought that affects this case is not comparable, for example, to the recent improvements in DNA testing. DNA testing can completely exclude a convicted defendant from the group of humans who could have committed the crime. CBLA normally allowed for testimony that a bullet involved in a crime was consistent with other bullets in the possession of the defendant, but those bullets may have come from a manufacturing batch that may have contained thousands of additional comparable bullets. In other words, CBLA allowed for circumstantial evidence suggesting a connection between the defendant and the bullet involved in the crime. Thus, the recent shift in scientific thought may reduce the amount of circumstantial evidence connecting the defendant to the offense, but it should not result in any affirmative evidence exonerating the convicted defendant.

I am not completely convinced that the normal test used for newly discovered evidence, i.e. that it would probably change the result, should be the test controlling this postconviction issue. See, e.g., Jones v. State, 709 So.2d 512, 521 (Fla.1998). Especially when the change in scientific thought occurs many years after the conviction, I am inclined to believe that the defendant should be required to establish a higher degree of certainty that the change in evidence would have altered the jury’s verdict. At this point, however, I do not believe that this case is ripe to resolve these difficult issues. These issues are sufficiently complex that it might be advisable for the trial court to appoint Mr. Smith counsel to assist him on remand.